UBC MDS

The Purpose of an Education Isn’t What You Think it Is

info-mds@science.ubc.ca — Mon, 26 Aug 2024 00:00:00 +0000

Doing homework and passing exams teach you more than just the course materials, but they also train you to be a motivated problem-solver.

“One day, I’ll work at AbCellera”, I told my partner. I didn’t want to spend my entire career working in a medical laboratory, performing repetitive tasks, and being stuck in a routine. I wanted more variety in my work, a chance to analyze and troubleshoot complex issues, and contribute to innovations that could make a real impact—whether through discovering new treatments, helping develop vaccines, or being part of a team that pushes the boundaries of science. That was my dream back in 2020, a year after graduating from the University of British Columbia (UBC) with a Bachelor of Science in microbiology, and shortly after I started working at Vancouver General Hospital. The excitement of being in the field I studied and the opportunity to contribute meaningfully to medical science energized me, and I started to envision a future with greater professional fulfillment and growth.

N-of-1 Trials and Analyzing Your Own Fitness Data

info-mds@science.ubc.ca — Thu, 15 Aug 2024 00:00:00 +0000

Photo by Luke Chesser on Unsplash.

I first heard of N-of-1 trials in 2018 as a master’s student studying epidemiology. I was in my Intermediate Epidemiologic and Clinical Research Methods class, and we had a guest lecture from Dr. Eric Daza on N-of-1 study design. The N-of-1 study can be thought of as a clinical trial investigating the efficacy of an intervention on an individual patient. At the time, this methodology was an emerging practice, with promising implications for personalized medicine and optimizing healthcare for the individual.

My Little Markov Model - Now Tweeting New Taylor Swift Lyrics

info-mds@science.ubc.ca — Thu, 01 Dec 2022 00:00:00 +0000

Picture This…

You’re sitting at your desk browsing the news and you see a head line something like:

(this is a real headline from Nov 2022)

You’re skeptical but you check out the AI startups demos and sure enough it looks like it could pass most first year university degrees with a 4.0.

Now imagine you’re in a Masters of Data Science program, working on a lab that is writing a basic text generation model from scratch and you read the same headline. I suspect you would feel somewhere about here on the Dunning-Kruger curve:

MDS Capstone via Zoom: A lookback

info-mds@science.ubc.ca — Mon, 12 Jul 2021 00:00:00 +0000

This post aims to share some of our learnings from our MDS Capstone project in 2021. If you’re a new student or a current student reading this after completing your block 6 coursework, congratulations! We wish you the very best in your capstone journey and hope that some of the following tips might serve you well during your project.

Project Management

Have a realistic timeline, give a buffer for your intended timeline

8 weeks for capstone projects is double the time of your earlier group projects in MDS, and you may think you have more time. However, more often than not, people tend to overestimate their productivity. Furthermore, from our experience, the first week for the hackathon and the last two weeks for presentation, testing & documentation went by much quicker than we realized, hence the core time for the project development was actually around 5 weeks.

Generating Photo-Realistic Neighbourhoods using Artificial Intelligence

info-mds@science.ubc.ca — Fri, 10 Jul 2020 00:00:00 +0000

Before starting, I wanted to give a brief introduction to the project, as well as give a special thanks to the team I worked with. This project was done for realtor.com as a UBC MDS capstone project. The team – Braden Tam, Florence Wang, Hanying Zhang and me – was given the task of using image generation to generate realistic pictures of neighbourhoods. The solution will address the issue of having empty thumbnails for some specific pages of the website. It is also worth mentioning that the images are not related to house postings (because we cannot sell a house that does not exist), but it is more related to the aesthetics of the website. Now that we know what the project is about, let’s start.

NLP in the Real World: A Reflection on a Two-Month Capstone

info-mds@science.ubc.ca — Fri, 10 Jul 2020 00:00:00 +0000

Recently, my team and I wrapped up our two-month long capstone project with Unbounce. The completion of the capstone marks the end of my Master of Data Science — Computational Linguistics program (MDS-CL) at the University of British Columbia, as well as the transition from being a student, the encapsulation of my identity for the past seventeen years, to being a new graduate, a new identity that still takes a little getting used to. In between job hunting and trying to maintain a semblance of a normal life in the midst of a global pandemic, I’d like to reflect on these past two months and the lessons that I’ve gained.

MDS 2019/2020 Capstone Seminar Series

info-mds@science.ubc.ca — Fri, 03 Jul 2020 00:00:00 +0000

During the last two months of the UBC Master of Data Science program (typically May & June each year) we have Capstone projects, in which our students work in teams of ~4 students with an external capstone partner and a UBC mentor to address a question facing the capstone partner’s organization using data science. In 2019 we added in a weekly “capstone seminar series” where we invite data science experts from academia and industry to give a seminar on a relevant data science topic. In this blog post I’ll provide a short description and some key take-aways from each of this year’s seminars. The schedule for the 2020 capstone seminar series is shown in Table 1 below.

Integrating R & Python into a Data Science program

info-mds@science.ubc.ca — Mon, 03 Feb 2020 00:00:00 +0000

R and Python are the two leading languages used in industry and academia for data analysis. Thus, to best prepare students in the University of British Columbia’s course-based, professional Master of Data Science (MDS) program to be competitive and perform on the job market, we have made an explicit decision to teach both languages.

Students are concurrently introduced to R and Python in the first month of MDS, and they work in both languages for all 8 months of their coursework. In some courses, students learn the same skills in both languages (e.g., general programming, data wrangling, visualization, software development and packaging) and in other courses we specialize in one language in order to go more deeply into that topic (e.g., statistical inference in R, and machine learning in Python). In the 2-month Capstone project at the end of the program, students work with a partner from industry, government or not-for-profit. At the start of the project, the students and partner agree on which programming language will be used. This decision is sometimes based on what best suits the project, and other times is chosen based on what the Capstone partner’s organization already uses. Other times the projects involve a mix of both languages.

Project courses in MDS

info-mds@science.ubc.ca — Thu, 22 Aug 2019 00:00:00 +0000

How can we train effective data scientists? Traditional lecture/lab-based courses typically involve prescribed and well-defined examples, and we found this format very effective for foundational courses that focus on a particular area of statistics, machine learning or computer programming. However, real-world data science differs greatly from these courses: data is messy, the tasks are not perfectly defined, knowledge must be integrated from various areas of data science, and collaboration with others is typically required. Our Master of Data Science (MDS) program ends with an 8-week full-time Capstone course, which provides much of the needed hands-on experience. However, based on student and instructor feedback, it was evident that we needed to do more. Thus, to augment and prepare students for Capstone, we have started transforming some of our traditional lecture/lab-based courses into smaller project-based courses as well.

What's for dinner? Predicting customer order probabilities

info-mds@science.ubc.ca — Fri, 26 Jul 2019 00:00:00 +0000

One of the things that drew to me data science is its applicability to pretty much any field you can name: technology, healthcare, finance, retail, education, government, entertainment, agriculture, real estate, etc. There’s no domain too large or small and no organization that would not benefit from having a data scientist (or a team of data scientists!) on staff to solve interesting problems.

I recently completed the Master of Data Science program at the University of British Columbia, a 10-month intensive program focused on computing, statistics, and machine learning. After 8 months of coursework, the program concludes with an 2-month capstone project. My capstone team worked with Vancouver-based meal kit company Fresh Prep to build a predictive model for customer ordering. Fresh Prep can use our model to understand which of their customers are likely to order in a given week, and target their marketing strategies in an attempt to increase order rates.

Winning the EasyMarkit AI Hackathon

info-mds@science.ubc.ca — Fri, 21 Jun 2019 00:00:00 +0000

On April 6, 2019, EasyMarkit hosted their first Hackathon in Vancouver where teams were asked to offer an AI solution to improve patient communication. My team (Bailey Lei, Alex Pak, Betty Zhou) was awarded first place based on the accuracy of our model in predicting communication response from patients.

About the EasyMarkit AI Hackathon

EasyMarkit is a Vancouver-based company that focuses on developing automated patient communication software specializing in dentistry. The hackathon challenged participants to predict whether or not patients will respond to communications like appointment reminders. The event was hosted at EasyMarkit’s office in downtown Vancouver, and ran from 9am to 5pm. Owen Ingraham, CTO at EasyMarkit, and his team did a fantastic job of hosting the event and our team was extremely impressed by the culture at EasyMarkit.

Teaching Convolutional Neural Networks

info-mds@science.ubc.ca — Mon, 15 Apr 2019 00:00:00 +0000

When I first learned about convolutional neural networks (also known as CNNs, or convnets), I was shown a picture much like the one below, which is from the AlexNet paper:

I will call this a “row-of-boxes” diagram. In my experience, this type of diagram is common in both CNN papers and CNN lessons, even though the audiences are very different in those two contexts. I would argue that row-of-boxes diagrams are targeted at people who already understand CNNs, not students seeing them for the first time. Given the diagram issue and other obstacles, it took me a few iterations to fully understand CNNs when I was a student.

Designing a Master of Data Science program: goals, design decisions, and lessons learned

info-mds@science.ubc.ca — Tue, 19 Feb 2019 00:00:00 +0000

Since launching the UBC MDS program in 2016, we’ve received a lot of questions on why we designed MDS the way we did. The post will address the following design decisions:

Statistics and CS as the home departments.
Goal of the program: responsible use of DS.
Length of the program: 10 months.
Length of the courses: 4 weeks.
Creating all new courses from scratch.
The program prerequisites.
Dividing the instructor role into two pieces: lecture and lab.
Setting a single deadline for all weekly assignments.
A few words on the curriculum.

Stat-CS partnership

The UBC MDS program is an equal partnership between the Department of Statistics and the Department of Computer Science. I have noticed that some other DS programs lack the statistics component, and I think that is a big loss. To us, data science is not just computer programming and machine learning (and certainly not just deep learning!). Rather, it is a broad field about how to ask and answer questions using data.

Welcome to our 2018-19 cohort

info-mds@science.ubc.ca — Mon, 17 Sep 2018 00:00:00 +0000

Last week we welcomed our third MDS Vancouver cohort. It only seems a short while ago when we launched the program with a cohort of “only” 22 students. Now we have 70 new students in the program with another 28 at our UBC Okanagan campus.

One of the strengths of the MDS Vancouver program is the diversity of our students:

70 new students.
41 domestic and 29 international students from Brazil, China, Hong Kong, Ivory Coast, India, Mexico, Puerto Rico, USA, Turkey and Pakistan.
32 women and 38 men (46% women, 54% men).
a variety of academic backgrounds, from psychology, neuroscience, business, economics, political science, biology, chemistry, engineering and many more.
Nine students have completed another advanced degree (Master’s, PhD, MD) prior to MDS.
60% of the class completed their Bachelor’s degree in 2016 or earlier, with 29% having just graduated in 2018.

We were extremely impressed with the quality and size of the applicant pool this year. The program is becoming more and more competitive but with more spaces opening up at our UBC Okanagan campus as well as a new MDS program with a specialization in Computational Linguistics coming in September 2019 (pending provincial government approval) there will be more opportunities for applicants to join our MDS family.

Bringing data science to new industries

info-mds@science.ubc.ca — Thu, 09 Aug 2018 00:00:00 +0000

Before starting the Master of Data Science (MDS) program at UBC, I had been working as a civil engineer for a consulting firm in Vancouver. I soon realized that despite producing vast amounts of data, the civil engineering and construction industries have felt little influence from advancements in data science and machine learning. Having seen the capabilities of data science and machine learning implemented in other industries, it became apparent that the civil engineering and construction industries could benefit greatly through the implementation of these technologies.

Learning from real data

info-mds@science.ubc.ca — Sat, 21 Jul 2018 00:00:00 +0000

For the past three months, I was on a team of UBC Master of Data Science (MDS) students working on our MDS capstone project. Our team developed a data science product to help the course creators on the Thinkific e-learning platform to improve their online courses. Unsurprisingly, working on the improvement of online learning was a great learning experience for ourselves. Therefore, to share some of my experiences, and as a reminder to my future self, in this post I will summarize the six most important lessons I learned during the project. But first, to give some context, I’ll quickly describe the client and the project.

Becoming a data scientist: My year-long hiatus from medical school

info-mds@science.ubc.ca — Wed, 04 Jul 2018 00:00:00 +0000

In my second year of medical school, I decided to open an unfamiliar email entitled “UBC Centennial Symposium on Health Informatics”. What was Health Informatics? I had no idea, but I intended to find out. The symposium featured the UK’s National Health Service (NHS) and showcased their use of large population datasets. I was impressed with how the NHS uses data to improve care. I went into medicine because of a personal desire to help people one-on-one; applying data science to healthcare seemed to provide the opportunity to positively impact millions at a time, in addition to one at a time.

Our curriculum, Part 1: Computer science & machine learning

info-mds@science.ubc.ca — Mon, 04 Jun 2018 00:00:00 +0000

This is the first in what will hopefully become a series of posts on our curriculum for the Master of Data Science (MDS) program at UBC. Our program is structured as six four-week blocks, each containing four mini-courses, for a total of 24 mini-courses. Each of these mini-courses is about one third the size of a regular university course. The image below summarizes our current schedule:

(For more details on each of these courses, see here.)

Visualizing massive open online courses

info-mds@science.ubc.ca — Mon, 01 Jan 2018 00:00:00 +0000

After eight months of coursework, the UBC Master of Data Science (MDS) program concludes with a 2-month Capstone project. We partnered with Ido Roll from UBC’s Centre for Teaching, Learning and Technology (CTLT) to analyze data from UBC’s massive open online courses (MOOCs). UBC offers dozens of MOOCs to thousands of students through the edX platform. Unlike in-person classes, in a MOOC the instructor cannot observe student engagement directly. Instead, they must infer student engagement from recorded events (for example, a student pausing a video). A single course can involve millions of events in its lifetime, which means there was lots of data for us to look through. Our task was to build a dashboard to help MOOC instructors answer the following questions:

Statistics-ML dictionary

info-mds@science.ubc.ca — Thu, 14 Dec 2017 00:00:00 +0000

One of the most rewarding aspects of working on the UBC Master of Data Science program has been the close collaboration between my home department, computer science, and the statistics department here at UBC. The collaboration has also come with a challenge, though: the two communities often use different words to mean the same thing. A prime example would be the word “bias” – it is hard to get an unbiased opinion on what this word means! Over the last couple of years I recorded these discrepancies in a “statistics / machine learning dictionary” and also expanded the scope of the document to include general terminology issues that arise in data science. The current version of the document can be found here. As one highlight, I’ve found that all of the following terms can refer to the same thing: predictors, features, inputs, explanatory variables, regressors, covariates, and independent variables! In general, terminology is a tricky business because there aren’t always objective truths and yet people tend to feel quite passionate about it.

Communication in data science: more than just the final report

info-mds@science.ubc.ca — Fri, 10 Nov 2017 00:00:00 +0000

When people stress the importance of good communication in data science, they are usually saying something like what Hadley Wickham says in his book R for Data Science: “[It] doesn’t matter how great your analysis is unless you can explain it to others: you need to communicate your results.” In other words, good communication is crucial to data science because it is the final step, without which all prior work would be wasted.

Algorithms and optimization

info-mds@science.ubc.ca — Wed, 18 Oct 2017 00:00:00 +0000

Over the years I’ve struggled with the disconnect between “algorithms” — as a student might see in a standard algorithms and data structures class — and optimization. Several of the algorithms taught in such courses are in fact instances of (discrete) optimization: for example, dynamic programming (DP), or Dijkstra’s algorithm for finding the shortest path in a graph (related to DP), or greedy algorithms for solving the traveling salesman problem. I like the example of dynamic programming in particular because it can often form a conceptual island in students’ minds, without apparent connection to any of their other knowledge (for example, in the famous CLRS algorithms book, DP is listed under the miscellaneous heading “Advanced Design and Analysis Techniques”). In this post I’ll do a deep dive into one of our MDS assignments (or “labs” as we call them — but they’re really assignments) that attempts to connect these concepts using a single application. The lab is available here, as part of our publicly available teaching materials.

Introducing our 2017-18 cohort

info-mds@science.ubc.ca — Fri, 01 Sep 2017 00:00:00 +0000

This is an exciting time for the UBC MDS program, with the second cohort starting their journey in just a few days. We are thrilled about the strength and diversity of our incoming class. Here are some facts about the incoming cohort:

43 new students.
21 domestic, 22 international.
20 women and 23 men (47% women, 53% men).
a variety of academic backgrounds, from psychology to civil engineering to a current UBC medical student who is taking a year out to pursue MDS.
previous degrees from UBC, other Canadian universities, the U.S., China, Colombia, France, India, South Africa, Turkey, and Vietnam (see map below).

Teaching with GitHub

info-mds@science.ubc.ca — Thu, 24 Aug 2017 00:00:00 +0000

In this post I will describe the system we use for delivering courses via GitHub in the UBC MDS program. I will first describe our high level goals and how we tackled them, and then discuss some implementation details. This post assumes the reader is familiar with git and GitHub and focuses more on the technical setup than the pedagogical implications.

Our main goals for this project were as follows:

help students gain experience with version control systems
eliminate all paper submissions
facilitate peer review amongst the students
facilitate group projects/assignments
allow course staff to view and comment on work-in-progress remotely
have a single place where most/all course content lives

When we were starting to think this through (around May 2016) there were already several existing solutions, such as GitHub Classroom, but none met all of our needs. We were committed to using GitHub rather than GitLab/Stash/etc because one of our team members, Jenny Bryan, had already been using GitHub to deliver STAT 545 and we wanted to build on her experience. The general approach is that students submit their work via GitHub and the TAs view/grade it there as well.

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Postdoctoral Teaching & Learning Fellow

The University of British Columbia, Vancouver invites applications for a Postdoctoral Teaching & Learning Fellow, associated with the Master of Data Science (MDS) program. This program is a collaborative effort of the Department of Computer Science and the Department of Statistics, within the Faculty of Science. The Fellow will be based in the Department of Statistics but work closely with colleagues from both departments.

The MDS program focusses on the innovative and responsible use of Data Science tools across a broad spectrum of data types and domain areas. It is a 10-month, full-time program, delivered in course modules of two-week and four-week durations. The last two months are devoted to a capstone project. The first cohort of students began their studies in September 2016.

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

This folder contains the images embedded in the with-great-data-science.md file in the _posts/ folder.

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

UBC MDS self-test

How to use this document

If you are a prospective or entering MDS student and you are not sure about your preparation, try the questions below. After some review of topics not recently encountered, a prepared student should be able to do all of the questions below with limited effort in 1-3 hours. These exercises are not meant to represent the level of MDS content (which will be much higher!); rather, the intent is that if you cannot do these questions easily, you are likely to struggle in the program without additional preparation. To acquire this preparation, please see our page of suggested learning resources.

About UBC MDS

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

The UBC Master of Data Science (MDS) program is a 10-month professional master’s program in Data Science. The University of British Columbia (UBC) is a comprehensive research-intensive university, consistently ranked among the 40 best universities in the world. The MDS program was launched in September 2016 and is offered by a collaboration between the UBC Department of Computer Science and Department of Statistics. The program involves 24 one-month courses followed by a two-month Capstone Project. MDS started at UBC’s main campus in Vancouver, and has since expanded to the Okanagan campus and to a computational linguistics option of the program. At UBC, our students learn to:

Application Tips

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

The following tips are meant to help prospective students prepare strong applications to the UBC MDS program.

1. Understand the program

Read about the program thoroughly, to fully understand what kind of students the program is looking for. MDS is a professional Master’s program, not a research program. We expect most of our graduates to enter the workforce after the program. While we also anticipate a few students may pursue further graduate studies in a domain area of interest, our program is not geared towards preparing students for further graduate studies in statistics or computer science. If your ultimate goal is to pursue a PhD in statistics or computer science, MDS is likely not the best fit for you.

Capstone Projects

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

The call for MDS-V Capstone project proposals is Now open.

During the last two months of the Master of Data Science program (May & June each year), our students work in teams of ~4 students with an external capstone partner and a UBC mentor to address a question facing the capstone partner’s organization using data science. The capstone program is free for partner organizations to participate in, but requires a point person who is willing to meet regularly with the students and offer guidance.

Career and Industry Resources

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Career and Professional Development

Build My Career UBC CareersOnline (exclusive job postings for UBC students and alumni - CWL required), resume, cover letter, interview skills, etc. resources and career and professional development events.
Grad Studies e-Resources and Presentations Career development and advice for UBC Graduate Students.
Mitacs Canadian research and training internships, career and professional development workshops and industry partnerships.
Computer Science Career Postings UBC Department of Computer Science events and employer information sessions (CWL required).
Glassdoor Interview questions-lookup for companies and roles.
Careercup Technical interview questions.
CodeEval Technical interview questions.
Programming Interviews Exposed Book on Amazon. Great for any Technical interviews.
Cracking the Coding Interview Book on Amazon. Lots of Technical interview questions.

Salary Information

Data Science

Vancouver Meetups

Learn Data Science and Data Science meetup groups
Python Group
R Group
Data Visualization Group
R-Ladies Vancouver

Facebook

Newsletters

Associations

Industry

Code of Conduct

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

MDS Leadership Code of Conduct

We include this code of conduct because we want to treat students respectfully.

Our Pledge

In the interest of fostering an open and welcoming environment, we as leaders of the Master of Data Science (MDS) program pledge to making studying under MDS a harassment-free experience for everyone, regardless of age, body size, disability, ethnicity, gender identity and expression, level of experience, education, socio-economic status, nationality, personal appearance, race, religion, or sexual identity and orientation. We pledge to abide by UBC’s Respectful Environment Statement.

Contributing to the MDS Blog

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Want to share your insights with the MDS community? Contributing to the MDS blog is easy and fun! Just follow these steps to get your blog post published with the help of our Editor, Varada Kolhatkar (@kvarada):

1. Reach Out

First things first—get in touch with the Editor, Varada Kolhatkar, at kvarada@cs.ubc.ca to discuss your blog idea and how you’d like to contribute. We’re excited to hear what you have in mind!

Data science terminology

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

About this document

This document is intended to help students navigate the large amount of jargon, terminology, and acronyms encountered in the MDS program and beyond. There is also an accompanying blog post. Course numbers of the form DSCI XXX refer to relevant UBC MDS courses.

Stat-ML dictionary

This section covers terms that have different meanings in different contexts, specifically statistics vs. machine learning (ML).

`regression`

in ML, regression refers to predicting continuous outputs given input features, and classification refers to predicting categorical outputs given input features.
in statistics, both of the above tasks are referred to as regression. See also supervised learning below.

`bias`

even within statistics this word has a lot of meanings. See also bias of an estimator.
in ML, when we have a trasformation of the form Wx+b (especially in linear models or neural networks) we refer to b as the “bias term” or the elements of b as the biases if b is a vector. For example, see this Stack Overflow post. In statistics we would call this the “intercept”.
in both fields, we talk about the bias-variance tradeoff; see below.

`parameter` and `parametric`

in statistics, parameter is used to describe probability distributions, like “the gamma distribution has a shape parameter and a scale parameter”. Thus, a parametric model is a model using a parametric probability distribution.
in ML, parameter refers to the components (usually numbers) that are getting learned in a system. A parametric model has a fixed number of parameters that is independent of the number of training examples, and typically doesn’t require the training examples to be stored in order to make predictions. An example would be linear regression, which involves one parameter per feature plus one more intercept parameter. On the other hand, k-nearest neighbours (KNN) would be an example of a nonparametric model as we don’t “distill” the training data into a fixed set of parameters. Another way to think about this is that, with KNN, the complexity of the model grows with the amount of training data.
the differences above can cause confusion. For example, a statistician might say a linear support vector machine (SVM) is not a parametric classifier because it is not based on an underlying probabilistic model. And yet, in ML, a linear SVM is parametric because we’re learning one parameter per dimension (in the primal formulation) to represent a linear boundary, and thus the number of parameters is fixed.

`early stopping`

in ML, this refers to terminating an optimization routine before reaching convergence, which may mitigate overfitting (see here).
in statistics, this sometimes refers to stopping an experiment early, particularly as in early stopping of clinical trials.

`hypothesis`

in statistics, this word evokes hypothesis testing.
in ML, hypothesis is sometimes used to refer to a particular model or decision boundary from the hypothesis space. E.g., we select a linear decision boundary from the hypothesis space of all possible hyperplanes; more complicated models have a larger hypothesis space.

`factor`

in statistics, factor means categorical variable; factorial experiment means trying all possible combinations of two or more factors.
spanning both statistics and ML, we have factor analysis and factor graphs, which do not use factor in the sense described above (or in the same way as each other, even…).
(and in math, factor means one of several things being multiplied together.)

`kernel`

This is not a Stat vs. ML problem, but more of a Stat/ML vs. CS vs. math problem. For even more definitions, see the Wikipedia disambiguation page. Overall, this word ranks up there with bias as a hopelessly overloaded and confusing word.

Dates and Deadlines

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

The MDS Mentoring Program is from November to April.

Deadline or Event	Date
Application Deadline	October 16, 2026
Kick-off Event	Week of November 16–20, 2026
Wrap-up Event	March 30, 2027

Contacts

MDS Mentoring Coordinators:

Milad Maymay, Director, Program Operations & Student Management
- Email: maymay@science.ubc.ca
Angela Pau, Career Advisor
- Email: angela.pau@ubc.ca

DSCI 511

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

Overview of data structures, iteration, flow control, and program design relevant to data exploration and analysis. When and how to exploit pre-existing libraries.

Learning Outcomes

By the end of the course, students are expected to be able to:

Write pseudo-code to specify, break down, and solve problems before being translated into code.
Write modular, easy-to-understand Python and R code that uses flow control, iteration, lists (arrays), and functions–and has appropriate style and organization.
Design and write Python and R programs to: perform calculations; read and write files; and use classes, objects, methods, and Python and R libraries. Determine which language (Python or R) is more appropriate for a given task.

Reference Material

Sedgewick, Robert; Wayne, Kevin; and Dondero, Robert. Introduction to Programming in Python: An Interdisciplinary Approach. Addison-Wesley, 2015.

DSCI 512

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

How to choose and use appropriate algorithms and data structures to help solve data science problems. Key concepts such as recursion and algorithmic complexity (e.g., efficiency, scalability).

Course Learning Objectives

By the end of the course, students are expected to be able to:

Apply fundamental algorithms such as sorting and searching, including iterative and recursive algorithms, using lists.
Select and justify the use of elementary data structures such as arrays, hash tables, trees, and simple graphs.
Analyze the scalability and trade-offs of various basic algorithms and data structures, using Big-O notation.
Explain why using a different (better) algorithm for a problem can result in a much, much bigger performance improvement than tweaking the algorithm already being used.
Apply basic discrete optimization methods such as dynamic programming.

Reference Material

Instructor (2016-2017)

Patrice Belleville

Note: information on this page is preliminary and subject to change.

DSCI 513

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

How to work with data stored in relational database systems or in formats utilizing markup languages. Storage structures and schemas, data relationships, and ways to query and aggregate such data.

Course Learning Objectives

By the end of the course, students are expected to be able to:

Use an Entity-Relationship Diagram and a database schema to determine the attributes, relationships, foreign keys, types of data, etc., in a given database.
Write SQL statements to query and update tables in a database.
Write programs containing embedded SQL statements to communicate with, and query, a database, thereby generating reports requiring more complicated logic and calculations than what is possible via stand-alone SQL.
Write well-formed XML; use XQuery to query an XML repository.

Reference Material

Ramakrishnan, Raghu and Gehrke, Johannes. Database Management Systems, 3rd Edition, McGraw-Hill, 2002.

Instructor (2016-2017)

Laks Lakshmanan

Note: information on this page is preliminary and subject to change.

DSCI 522

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

Interactive vs. scripted/unattended analyses and how to move fluidly between them. Reproducibility through automation and dynamic, literate documents. The use of version control and file organization to enhance machine- and human-readability.

Learning Outcomes

By the end of the course, students are expected to be able to:

Analyze data interactively using read-eval-print-loop (REPL) processes; write scripts for non-interactive use; use tools and work styles to create fluidity between these two modes (e.g., RStudio IDE, iPython).
Perform dynamic reporting functions such as integrating narrative, code, data, numerical results, and visual results; create reproducible reports and workflows (e.g., R Markdown, Project Jupyter).
Manage projects by designing workflows for self-documentation, reproducibility, and collaboration; organize files with appropriate naming conventions; manage paths and dependencies.
Use version control software (e.g., Git) including distributed version control and remote servers (e.g., GitHub, Bitbucket).
Automate data science workflows (using e.g., Make, Galaxy).

Prerequisites

DSCI 511 (Programming for Data Science)
DSCI 521 (Computing Platforms for Data Science)

Reference Material

TBD

DSCI 523

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

Converting data from the form in which it is collected to the form needed for analysis. How to clean, filter, arrange, aggregate, and transform diverse data types, e.g. strings, numbers, and date-times.

Learning Outcomes

By the end of the course, students are expected to be able to:

Analyze and determine appropriate ways of manipulating a single data table using various techniques including: filtering rows or observations based on a criterion or combination of criteria; selecting variables (columns); arranging observations or variables in a deliberate way (e.g., sorting, grouping); forming new variables from one or more existing variables; reshaping data; computing summaries on groups of observations based on one or more categorical variables.
Handle common and tricky data types; manipulate text, dates/times, strings, and regular expressions; detect and handle duplicates and outliers.
Determine appropriate manipulations for two-table data, including lookups and joins with suitably-selected columns.
Handle non-tabular data (e.g., nested lists) in the languages being used (e.g., Python, R); convert data among general formats (e.g., XML, JSON).

Prerequisites

DSCI 521 (Computing Platforms for Data Science)
DSCI 511 (Programming for Data Science)

Reference Material

Instructor (2016-2017)

Jenny Bryan

Note: information on this page is preliminary and subject to change.

DSCI 524

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

How to exploit practices from collaborative software development techniques in data scientific workflows. Appropriate use of abstraction and classes, the software life cycle, unit testing / continuous integration, and packaging for use by others.

Learning Outcomes

By the end of the course, students are expected to be able to:

Package code for use by others, including the specification of dependencies/requirements.
Use distributed version control and issue tracking to manage multi-person projects.
Specify, implment, and use a data abstraction in Python; write a comprehensive test suite for a data abstraction in Python.
Implement and call S3 methods in R; specify an object’s class in R
Handle exceptional cases in a function or method with exceptions or assert statements
Interpret software licenses and select software licenses that best suit the needs of software that they create.

Reference Material

Instructor (2016-2017)

Meghan Allen

Note: information on this page is preliminary and subject to change.

DSCI 525

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

How to use the web as a platform for data collection, computation, and publishing. Accessing data via scraping and APIs. Using the cloud for tasks that are beyond the capability of your local computing resources.

Learning Outcomes

By the end of the course, students are expected to be able to:

Scrape data from websites and access data using application programming interfaces (APIs) where available.
Author web content for public access.
Host a simple application on a cloud computing platform such as Amazon’s EC2.
Connect the concepts in databases to those of distributed computing
Parallelize computations in an environment such as iPython Parallel.

Reference Material

Instructor (2016-2017)

Mike Feeley

Note: information on this page is preliminary and subject to change.

DSCI 531

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

The design and implementation of static figures across all phases of data analysis, from ingest and cleaning to description and inference. How to make principled and effective choices with respect to marks, spatial arrangement, and colour.

Learning Outcomes

By the end of the course, students are expected to be able to:

Analyze existing static visual encodings in terms of marks and channels, spatial arrangement, and color broken down into luminance, saturation, and hue.
Design new static visual encodings that use space and color channels appropriately according to principles of perceptual effectiveness and by matching channel type to attribute type for quantitative versus categorical attributes.
Implement static visual encodings using existing toolkits and libraries.
Describe and manipulate table, network, and spatial data; transform data into a form suitable for the intended abstract task of the visualization user.
Explain whether a visual encoding is perceptually appropriate for a specific combination of task and data.

Reference Material

Munzner, Tamara. Visualization Analysis and Design, CRC Press, 2014.

Instructor (2016-2017)

Tamara Munzner

Note: information on this page is preliminary and subject to change.

DSCI 532

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

Analysis, design, and implementation of interactive figures. How to provide multiple views, deal with complexity, and make difficult decisions about data reduction.

Learning Outcomes

By the end of the course, students are expected to be able to:

Analyze interactive visualizations in terms of: dynamic change over time, partitioning into multiple views, and data reduction within a single view.
Design new interactive visualizations for complex datasets where a single static view is not sufficient for the intended analysis task.
Implement interactive visualizations using existing toolkits and libraries.
Explain the trade-offs of using 3D vs 2D representations and of using animation vs juxtaposed views.
Explain and justify methods to validate visualization design effectiveness including computational benchmarks, field studies on deployed software, and qualitative discussion of visual results.

Reference Material

Munzner, Tamara. Visualization Analysis and Design, CRC Press, 2014.

Instructor (2016-2017)

Tamara Munzner

Note: information on this page is preliminary and subject to change.

DSCI 541

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

The legal, ethical, and security issues concerning data, including aggregated data. Proactive compliance with rules and, in their absence, principles for the responsible management of sensitive data. Case studies.

Learning Outcomes

By the end of the course, students will be able to:

Identify situations in which data is sensitive, assess the risks, and articulate a reasoned response.
Identify the pros and cons of situations in which data was collected for one purpose and later analyzed for other purposes.
Explain trade-offs in security and privacy; explain why cybersecurity is complex and difficult.
Explain the purpose of an ethics board, determine when an ethics board needs to be consulted, and explain the various forms of consent. Consider privacy laws and legal issues.
Implement good security and privacy practices in data storage, use, and reporting.

Reference Material

Baase, Sara. A Gift of Fire: Social, Legal, and Ethical Issues in Computer Technology, 4th Edition. Pearson, 2012.
Rosenberg, Richard. The Social Impact of Computers, 4th Edition. Emerald Group, 2004.
Schneier, Bruce. Data and Goliath: The Hidden Battles to Collect Your Data and Control Your World. Norton, 2015.
Schneier, Bruce. Secrets and Lies: Digital Security in a Networked World, 15th Anniversary Edition. Wiley, 2015.
Free online resources, available on the Internet. For example:
- Abelson, Hal; Ledeen, Ken; and Lewis, Harry. Blown to Bits: Your Life, Liberty, and Happiness after the Digital Explosion. Addison-Wesley, 2008. http://www.bitsbook.com/excerpts/.
- Geist, Michael. http://www.michaelgeist.ca/blog/ (blog, privacy, ethics, Canadian focus)
- Krebs, Brian. https://krebsonsecurity.com/ (blog on security)
- Schneier, Bruce. Crypto-Gram Monthly Blog and Archives. https://www.schneier.com/crypto-gram/ (blog on security, privacy, ethics)

Instructor (2016-2017)

Ed Knorr

Note: information on this page is preliminary and subject to change.

DSCI 542

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

Effective oral and written communication, across diverse target audiences, to facilitate understanding and decision-making. How to present and interpret data, with productive skepticism and an awareness of assumptions and bias.

Learning Outcomes

By the end of the course, students are expected to be able to:

Outline the components of a good scientific argument, paying attention to claims, reasons, evidence, assumptions, bias, validity, reliability, etc.
Identify the components of a good experiment or data collection effort, paying attention to how the data was collected and how it is being used to construct a scientific model; identify limitations of the data and model.
Work effectively with teams and domain experts on data science problems.
Communicate uncertainty to diverse audiences.
Explain the purpose and strengths of consistent documentation practices.
Write effectively on technical data science topics for a nontechinal audience.
Present data science results to diverse audiences and recommend subsequent action to decision makers.
Communicate effectively through oral presentations and written reports. Distinguish between the goals of each.

Reference Material

Booth, Wayne; Colomb, Gregory; and Williams, Joseph. The Craft of Research, 3rd Edition, Chicago Guides to Writing, Editing, and Publishing, University of Chicago Press, 2008. (also available as a free download)
Aaron, Jane and Morrison, Aimee. The Little, Brown Compact Handbook, 5th Canadian Edition, Pearson, 2012.
Messenger, William E.; de Bruyn, Jan; Brown, Judy; and Montagnes, Ramona. The Canadian Writer’s Handbook, 6th Edition, Oxford University Press, 2014,
Reynolds, Garr. Presentation Zen: Simple Ideas on Presentation Design and Delivery, 2nd Edition, New Riders, 2011.
Zelazny, Gene. Say It with Charts, 4th Edtition, McGraw-Hill, 2001.

Instructor (2016-2017)

Ed Kroc

Note: information on this page is preliminary and subject to change.

DSCI 551

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

Describing data in terms of its location, spread, and general distribution. How to balance the use of procedures from classical, parametric statistics with robust approaches that account for outliers and missing data.

Learning Outcomes

By the end of the course, students will be able to:

Define probability concepts such as random variables, the distribution of a random variable, and parameters of a distribution.
Describe quantitative or continuous variables, and explain measures of location, spread, and other more complicated features of a distribution.
Describe the connection between probability concepts and observed data, including the distinction between a property of the true distribution and its empirical counterpart (e.g., true mean vs sample mean), measures of uncertainty (e.g., standard error), and interval estimation.
Compute descriptive statistics, both low dimensional (e.g., sample mean, variance, and median) and high dimensional (e.g., empirical distribution, histogram estimator, and kernel density estimator).
Describe categorical variables. Explain and compute relevant measures such as frequency, relative frequency, entropy, and mode.
Identify problems in observed data, such as contamination with outliers or missing data. Implement mitigation strategies, such as robust statistics and imputation.
Guard against biases potentially caused by missing data.

Reference Material

Introduction to Statistics Through Resampling Methods and R
Roger Peng. Exploratory Data Analysis with R. https://leanpub.com/exdata
Jeff Leek. The Elements of Data Analytic Style. https://leanpub.com/datastyle

Instructor (2016-2017)

Jenny Bryan

Note: information on this page is preliminary and subject to change.

DSCI 552

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

The statistical and probabilistic foundations of inference, developed jointly through mathematical derivations and simulation techniques. Important distributions and large sample results. The frequentist paradigm.

Learning Outcomes

By the end of the course, students are expected to be able to:

Manipulate the most important probability distributions, and correctly identify appropriate distributions for different modeling situations.
Design, perform and interpret frequentist hypothesis tests in the context of parameter estimation.
Work proficiently with standard statistical notation for parameters, sample quantities, estimators, etc.
Apply large sample results including the Law of Large Numbers and the Central Limit Theorem.
Design and perform appropriate simulation techniques to make predictions and understand the relationship between models and data.

Reference Material

Instructor (2016-2017)

Mike Marin

Note: information on this page is preliminary and subject to change.

DSCI 553

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

Methods for dealing with the multiple testing problem. Bayesian reasoning for data science. How to formulate and implement inference using the prior-to-posterior paradigm.

Learning Outcomes

By the end of the course, students will be able to:

Use Bayesian reasoning when modeling data.
Apply Bayesian statistics to regression models.
Compare and contrast Bayesian and frequentist methods, and evaluate their relative strengths.
Use appropriate statistical libraries and packages for performing Bayesian inference (e.g., PyMC).
Avoid the pitfalls of multiple comparisons by using the proper corrections.

Reference Material

Gelman, Andrew and Carlin, John B. Bayesian Data Analysis, 3rd Edition. Chapman and Hall, 2013.

Instructor (2016-2017)

Paul Gustafson

Note: information on this page is preliminary and subject to change.

DSCI 554

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

Statistical evidence from randomized experiments versus observational studies. Applications of randomization, e.g., A/B testing for website optimization.

Learning Outcomes

By the end of the course, students are expected to be able to:

Distinguish between experimentally-generated data and observational data, with particular reference to the strength of ensuing statistical conclusions.
Fit and interpret regression models for observational data, with particular reference to adjustment for potential confounding variables.
Apply the principle of “block what you can, randomize what you cannot” in designing an A/B testing experiment.
Choose appropriately between fixed-effect and random-effect regression models.

Reference Material

O’Neil, Cathy and Schutt, Rachel. “Causality,” Ch. 11 of Doing Data Science: Straight Talk from the Frontline, O’Reilly Media, 2013.
Tang, Diane, et al. “Overlapping Experiment Infrastructure: More, Better, Faster Experimentation.” Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, ACM, 2010.

Instructor (2016-2017)

Paul Gustafson

Note: information on this page is preliminary and subject to change.

DSCI 561

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

Linear models for a quantitative response variable, with multiple categorical and/or quantitative predictors. Matrix formulation of linear regression. Model assessment and prediction.

Learning Outcomes

By the end of the course, students are expected to be able to:

Fit and interpret a linear regression model.
Identify whether a linear regression model is appropriate for a given dataset.
Critique a specific regression model applied to a given dataset on the basis of both diagnostic plots and hypothesis tests.
Specify and interpret interaction terms and nonlinear terms.

Reference Material

Faraway, Julian J. Linear Models with R, 2nd Edition. Chapman and Hall, 2014.

Instructor (2016-2017)

Gabriela Cohen Freue

Note: information on this page is preliminary and subject to change.

DSCI 562

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

Useful extensions to basic regression, e.g., generalized linear models, mixed effects, smoothing, robust regression, and techniques for dealing with missing data.

Learning Outcomes

By the end of the course, students are expected to be able to:

Identify appropriate alternatives for problems where a linear regression model should not be used, and discuss the specific difficulties that are being addressed.
Discuss the advantages and disadvantages of using non-parametric regression methods.
Fit a mixed effects model when appropriate, and interpret the corresponding parameter estimates.
Correctly apply robust estimators to determine whether outliers are present in the data, and explain the implications of their removal on subsequent analyses of the data.

Reference Material

Maronna, Ricardo; Martin, Doug; and Yohai, Victor. Robust Statistics: Theory and Methods. Wiley, 2006.

Instructor (2016-2017)

Lang Wu

Note: information on this page is preliminary and subject to change.

DSCI 563

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

How to find groups and other structure in unlabeled, possibly high dimensional data. Dimension reduction for visualization and data analysis. Clustering, association rules, model fitting via the EM algorithm.

Learning Outcomes

By the end of the course, students are expected to be able to:

Explain, with examples, the key differences between a supervised and an unsupervised learning problem.
Apply successfully K-means and K-medoids, hierarchical and model-based clustering, including the EM algorithm.
Explain and apply appropriately the following dimension-reduction methods: principal components, factor analysis and multidimensional scaling. Explain their differences and similarities.
Explain and apply appropriately different matrix decompositions, including Singular Value Decomposition, Cholesky Decomposition, QR and LU.
Apply and correctly interpret relevant visualization tools to the analysis (e.g., heatmaps and dendrograms).

Instructor (2016-2017)

Matias Salibian-Barrera

Note: information on this page is preliminary and subject to change.

DSCI 571

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

Introduction to supervised machine learning, with a focus on classification. Decision trees, logistic regression, and basic machine learning concepts such as generalization error and overfitting.

Learning Outcomes

By the end of the course, students are expected to be able to:

Explain, with examples, the key differences between an unsupervised learning problem and a supervised learning problem, as well as the concept of training and test data.
Construct and apply a decision tree classification model, and explain how the concept of generalization error is tied to the depth of a decision tree.
Apply logistic regression to classification, and explain the key differences between regression and classification models.
Apply regularization in the context of logistic regression.
Build a k-th nearest-neighbor (kNN) classifier; compare and contrast parametric and non-parametric classification models.
Deploy support vector machine (SVM) classifiers, and explain how kernel functions are used in such classifiers.
Avoid the pitfalls of overfitting and reusing test sets.

Reference Material

Russell, Stuart, and Peter Norvig. Artificial intelligence: a modern approach. Third Edition. 2010.
David Poole and Alan Mackwordth. Artificial Intelligence: foundations of computational agents. 2010. (free online http://artint.info/)

Instructor (2016-2017)

Giuseppe Carenini

Note: information on this page is preliminary and subject to change.

DSCI 572

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

An introduction to optimization for machine learning. Computation of derivatives. Deep learning.

Learning Outcomes

By the end of the course, students are expected to be able to:

Formulate various machine learning problems as optimization problems.
Implement gradient descent and compare/contrast with stochastic gradient descent.
Avoid common numerical errors due to rounding error.
Compare/contrast different ways of computing derivatives (symbolic/automatic/numerical differentiation).
Train neural networks for performing regression and classification tasks with deep learning.
Deploy neural networks on a GPU.

Reference Material

Ian Goodfellow, Yoshua Bengio and Aaron Courville. http://www.deeplearningbook.org/
James, Gareth; Witten, Daniela; Hastie, Trevor; and Tibshirani, Robert. An Introduction to Statistical Learning: with Applications in R. 2014. Plus Python code and more Python code.
Russell, Stuart, and Peter Norvig. Artificial intelligence: a modern approach. 1995.
David Poole and Alan Mackwordth. Artificial Intelligence: foundations of computational agents. 2010.
Kevin Murphy. Machine Learning: A Probabilistic Perspective. 2012.
Christopher Bishop. Pattern Recognition and Machine Learning. 2007.
Pang-Ning Tan, Michael Steinbach, Vipin Kumar. Introduction to Data Mining. 2005.
Mining of Massive Datasets. Jure Leskovec, Anand Rajaraman, Jeffrey David Ullman. 2nd ed, 2014.
A Course in Machine Learning
Deep Learning with Python. Jason Brownlee.
Stanford UFLDL tutorial (or here)
Nando de Freitas lecture videos and online course
Neural Networks and Deep Learning (free online book)
Grokking Deep Learning
Practical Deep Learning For Coders, Part 1 and some more resources on their blog here
A Guide to Deep Learning
a list of other resources: https://github.com/ChristosChristofidis/awesome-deep-learning

Instructor (2016-2017)

Mike Gelbart

Note: information on this page is preliminary and subject to change.

DSCI 573

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

How to evaluate and select features and models. Cross-validation, ROC curves, feature engineering, the role of regularization. Automating these tasks with hyperparameter optimization.

Learning Outcomes

By the end of the course, students are expected to be able to:

Explain how the concepts of generalization error and overfitting of training data are essential to the performance of a classification or regression model.
Apply and use shrinkage and feature selection methods (e.g., Lasso, elastic nets).
Perform k-fold cross validation and bootstrapping based on the training data.
Evaluate the quality of a statistical model in order to do model/feature selection.
Explain how the ROC is generated, and how the area under the ROC curve (AUC) can be used for comparing models.
Diagnose/understand/address overfitting and underfitting.

Instructor (2016-2017)

Mark Schmidt

Note: information on this page is preliminary and subject to change.

DSCI 574

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

Model fitting and prediction in the presence of correlation due to temporal and/or spatial association. ARIMA models and Gaussian processes.

Learning Outcomes

By the end of the course, students are expected to be able to:

Explain, with examples, the main inferential tasks related to spatial or temporal data, and how the spatial or temporal associations make it possible to borrow statistical tools.
Understand the ideas of autocorrelation and correlated errors, and be able to explain the importance of these ideas for temporal and spatial modelling.
Fit temporal, spatial, and spatio-temporal models by implementing them in a probabilistic programming language, and interpret the results.
Apply relevant visualization tools and draw correct conclusions from the analysis.

Reference Material

Shaddick, Gavin and Zidek, James V. Spatio-Temporal Methods in Environmental Epidemiology. CRC Press, 2016.
Chatfield, Chris. The Analysis of Time Series: An Introduction. CRC Press, 2003.

Instructor (2016-2017)

Natalia Nolde

Note: information on this page is preliminary and subject to change.

DSCI 575

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

Advanced machine learning methods, with an undercurrent of natural language processing (NLP) applications. Bag of words, recommender systems, topic models, ranking, natural language as sequence data, POS tagging, CRFs for named entity recognition and RNNs for text synthesis. An introduction to popular NLP libraries in Python.

Learning Outcomes

By the end of the course, students are expected to be able to:

Compare and contrast classifiers that generate binary predictions and those that compute probabilistic predictions; derive the probabilistic predictions of the Naive Bayes method.
Apply graphical models as a probabilistic approach to model complex, large-scale problems.
Apply basic techniques in active data acquisition, and explain under what circumstances these techniques are worth using.
Make use of pairwise preference data via ranking algorithms
Identify when recommender systems may be useful and apply them in these circumstances.

Instructor (2016-2017)

Mark Schmidt

Note: information on this page is preliminary and subject to change.

DSCI 591

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

A mentored group project based on real data and questions from a partner within or outside the university. Students will formulate questions and design and execute a suitable analysis plan. The group will work collaboratively to produce a project report, presentation, and possibly other products, such as a web application.

Learning Outcomes

By the end of the course, students are expected to be able to:

Identify an interesting data science question for which data is available or obtainable.
Define the scope of a possible solution, identify units of work (deliverables), and estimate the effort required.
Design and implement a solution to a problem in data science that can be completed within 8 weeks.
Function effectively in teams: communicate productively between team members, identify sub-problems that could be worked on individually by team members, and integrate contributions of team members into a final product.
Document and present (using written, oral, and visual means) the process and results from a solution to a data science problem.
Evaluate or assess a solution to a data science problem, and compare it with alternative approaches.

DSCI-521

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Short Description

How to install, maintain, and use the data scientific software “stack”. The Unix operating system, integrated development environments, and problem solving strategies.

Learning Outcomes

By the end of the course, students are expected to be able to:

Customize and configure software platforms used in the MDS program.
Employ appropriate computing concepts and state-of-the-art Integrated Development Environments (IDEs) in the programming process.
Diagnose and troubleshoot programming and development environment problems, and explain how such problems can be avoided.
Integrate popular R and Python libraries into their code; select appropriate libraries for a given task.

Reference Material

UNIX Power Tools Jerry Peek, Tim O’Reilly & Mike Loukides

Instructors (2016-2017)

Note: information on this page is preliminary and subject to change.

General lab instructions

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Lab section and groups

Students are assigned to one of the two lecture sections, 001 and 002, before course registration. The lecture section does not change throughout the program and we do not accept requests to change sections (See the policy here). Each lecture section is divided into two lab sections: students in Section 001 are assigned to either L01 or L02, while students in Section 002 are assigned to either L03 or L04.

Installation Guide

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

These instructions will walk you through how to install the required Data Science software stack for the UBC Master of Data Science program. Before starting with the installation instructions, ensure that your laptop meets our program requirements. Students’ whose laptops do not meet the requirements specified below will not be able to receive technical assistance from the MDS team in troubleshooting installation issues.

Laptop requirements

Runs one of the following operating systems: Ubuntu 22.04 (any version >=20.04 will likely work), macOS Big Sur or above (>= 11.4.x), Windows 11 Professional, Enterprise or Education (version 2004, 20H2, or 21H1).
- When installing Ubuntu, checking the box “Install third party…” will (among other things) install proprietary drivers, which can be helpful for wifi and graphics cards.
Can connect to networks via a wireless connection for on campus work
Has access to an internet connection that is fast and stable enough for video calling and conducting online quizzes
Has at least 50 GB disk space available
Has at least 8 GB of RAM
Uses a 64-bit CPU
- M series Apple laptops are supported with workarounds
- Windows on ARM is not compatible
Is at most 6 years old at the start of the program (4 years old or newer is recommended)
Uses English as the default language
Student user has full administrative access to the computer

Software installation instructions

Please click the appropriate link below to view the installation instructions for your operating system:

Jobs

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Current Positions

Call for TAs for 2026W1 - Closed

Learning Resources

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

The following is a list of online resources that might be useful preparation for the UBC MDS program. Completion of these courses does not replace the official program prerequisites. Rather, this page is mainly intended for entering students who may wish to reinforce their preparation before the program starts. Many of these resources can also be helpful resources during the program. Highly recommended items are in bold.

Disclaimer: We have not vetted all these resources ourselves, but rather selected them based on a combination of our experience, recommendation from colleagues, and the resources descriptions. If you have feedback about them, please let us know.

Legal

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Background

Some MDS-V Capstone projects may require agreements to be put in place to protect confidential background information and to define how new project intellectual property, if any is developed, will be handled. We ask Capstone partners to disclose if their project falls under this category in their project proposal so that students are aware of this requirement if they choose to work on the project. For all MDS-V Capstone projects that have such requirements, the projects will be bound by the terms of the agreements outlined below.

macOS

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Installation notes
UBC Student Email
Web browser
Password manager
Slack
Bash shell
Xcode command line tools
Visual Studio Code
GitHub
Git
Python, Conda, and JupyterLab
R, XQuartz, IRkernel, and RStudio
Quarto CLI
LaTeX
PostgreSQL
Docker
VS Code extensions
Improving the bash configuration
Post-installation notes

Important Note that there are differences in some parts of the installation for Mac computers with recent Apple Silicon (Mac M1-M4) and earlier Intel chips. If you have a newer Mac laptop, make sure to chose relevant versions (usually denoted as Apple Silicon, Mac M1-M4, Mac arm64 or Darwin). For older Intel Macs, in all the sections below, if you are presented with the choice to download either a 64-bit (also called x64) or a 32-bit (also called x86) version of the application always choose the 64-bit version.

MDS Calendar

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Note: Only current MDS students, faculty, and staff members can view all the calendar contents. If you have any questions, please get in touch with the course coordinators.

Prospective students be aware that MDS typically follows a one week-shifted schedule to the UBC start dates. Please refer to our official website and email communications for your start and end dates, including any scheduled breaks.

Legend:

Blue: lectures
Green and orange: labs
Yellow: office hours and other academic activities
Purple: quizzes
Grey: deadlines
Red: extracurricular activities/events

How to subscribe to the calendars:

MDS Courses

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Course Number	Block	Course Title	Short Description	Expanded Description	Section 1 Instructor	Section 2 Instructor
DSCI 511	1	Programming for Data Science	Pseudo-code. Program design and structure. Flow control. Iteration. Lists (arrays). Functions. File I/O. Classes, objects, methods, and libraries.	Program design and data manipulation with Python. Overview of data structures, iteration, flow control, and program design relevant to data exploration and analysis. When and how to exploit pre-existing libraries.	Prajeet Bajpai	Elham E Khoda
DSCI 521	1	Computing Platforms for Data Science	Introduction to software, shells, tools, and file systems for use in the Data Science program. Installation, configuration, and use of statistical and programming software including Integrated Development Environments (IDEs). Problem resolution skills.	How to install, maintain, and use the data scientific software stack. The Unix shell, version control, and problem solving strategies. Literate programming documents.	Ilya Musabirov	Daniel Chen
DSCI 523	1	Programming for Data Manipulation	Program design and data manipulation using industry-standard software tools designed for statistical work. Organizing, filtering, sorting, grouping, reformatting, converting, and cleaning data to prepare it for further analysis.	Program design and data manipulation with R. Organizing, filtering, sorting, grouping, reformatting, converting, and cleaning data to prepare it for further analysis.	Gittu George	Tiffany Timbers
DSCI 551	1	Descriptive Statistics and Probability for Data Science	Descriptive statistics including measures of location and spread. Random variables, distributions, and parameters. Categorical variables. Uncertainty. Missing data.	Fundamental concepts in probability including conditional, joint, and marginal distributions. Statistical view of data coming from a probability distribution.	Payman Nickchi	Alexi Rodríguez-Arelis
DSCI 512	2	Algorithms and Data Structures	Basic algorithms. Recursion. Data structures including linked lists, queues, stacks, trees, graphs, and hash tables. Searching and sorting. Introduction to complexity including Big-O notation, efficiency, and scalability.	How to choose and use appropriate algorithms and data structures to help solve data science problems. Key concepts such as recursion and algorithmic complexity (e.g., efficiency, scalability).	Elham E Khoda	Sky Sheng
DSCI 531	2	Data Visualization I	Descriptive plots using statistical and programming software. Basics, mechanics, and principles of data visualization.	Exploratory data analysis. Design of effective static visualizations. Plotting tools in R and Python.	Joel Östblom	Joel Östblom
DSCI 552	2	Statistical Inference and Computation I	Random variables, parameters, observed data, statistics (distinctions and connections). Estimation: point and interval. Two-group comparisons, frequentist version. Simulation-based approaches.	The statistical and probabilistic foundations of inference. Large sample results. The frequentist paradigm.	Alexi Rodríguez-Arelis	Rodolfo Lourenzutti
DSCI 571	2	Supervised Learning I	Decision trees. k-th nearest neighbour classifiers. Naive Bayes classifiers. Logistic regression.	Introduction to supervised machine learning. Basic machine learning concepts such as generalization error and overfitting. Various approaches such as K-NN, decision trees, linear classifiers.	Varada Kolhatkar	Varada Kolhatkar
DSCI 513	3	Databases and Data Retrieval	Relational schemas. SQL queries. Database programming using embedded SQL. XML and XQuery.	How to work with data stored in relational database systems. Storage structures and schemas, data relationships, and ways to query and aggregate such data.	Gittu George	Gittu George
DSCI 522	3	Data Science Workflows	Interactive and non-interactive data analysis. Scripting. Dynamic reporting. Reproducibility. Project and file management. Version control. Automated workflows.	Interactive vs. scripted/unattended analyses and how to move fluidly between them. Reproducibility through automation and containerization.	Daniel Chen	Sky Sheng
DSCI 561	3	Regression I	Linear models: continuous response; one or more categorical covariates and/or one or more continuous covariates.	Linear models for a quantitative response variable, with multiple categorical and/or quantitative predictors. Matrix formulation of linear regression. Model assessment and prediction.	Rodolfo Lourenzutti	Payman Nickchi
DSCI 573	3	Feature and Model Selection	Performance of a classification model. Generalization error, overfitting of training data. Shrinkage, feature selection, Akaike Information Criterion, Bayesian Information Criterion. k-fold cross validation. Bootstrapping. Receiver Operating Characteristic curve. Elastic nets, regularization.	How to evaluate and select features and models. Cross-validation, ROC curves, feature engineering, and regularization.	Prajeet Bajpai	Elham E Khoda
DSCI 524	4	Collaborative Software Development	Software life cycle. Unit testing. Continuous integration. Submission to a relevant repository for distribution. Packaging for installation and use by others. Software licenses. Classes and abstraction.	How to exploit practices from collaborative software development techniques in data scientific workflows. Appropriate use of abstraction, the software life cycle, unit testing / continuous integration, and packaging for use by others.	Tiffany Timbers	Daniel Chen
DSCI 541	4	Privacy, Ethics, and Security	Privacy and data. Ethics boards, legal issues, licensing. Physical and logical data security, social engineering. Encryption, data anonymization, privacy-preserving techniques. Case studies.	The legal, ethical, and security issues concerning data, including aggregated data. Proactive compliance with rules and, in their absence, principles for the responsible management of sensitive data. Case studies.	Sky Sheng	Sky Sheng
DSCI 562	4	Regression II	Non-parametric regression and smoothing. Data-driven parameter selection. Robust regression. Mixed effects.	Useful extensions to basic regression, e.g., generalized linear models, mixed effects, smoothing, robust regression, and techniques for dealing with missing data.	Alexi Rodríguez-Arelis	Payman Nickchi
DSCI 572	4	Supervised Learning II	Support Vector Machines. Random Forests. Ensemble Classifiers. Graphical models.	Introduction to numerical optimization (e.g., gradient descent). Neural networks and deep learning.	Varada Kolhatkar	Prajeet Bajpai
DSCI 525	5	Web and Cloud Computing	Networks and the Internet, scraping data, APIs, cloud computing, Web services for scalable computing, Web hosting, Web publication platforms, introduction to parallel computing.	How to use the web as a platform for data collection, computation, and publishing. Accessing data via scraping and APIs. Using the cloud for tasks that are beyond the capability of your local computing resources.	Gittu George	Ilya Musabirov
DSCI 553	5	Statistical Inference and Computation II	Multiple hypothesis testing, false discovery rate. Two-group comparisons, Bayesian paradigm.	Bayesian reasoning for data science. How to formulate and implement inference using the prior-to-posterior paradigm.	Payman Nickchi	Alexi Rodríguez-Arelis
DSCI 563	5	Unsupervised Learning	Unsupervised learning. K-means/medoids. Model-based clustering. Expectation-maximization algorithm. Hierarchical clustering. Dimension reduction. Matrix decomposition. Heatmaps, contour plots, dendograms.	How to find groups and other structure in unlabeled, possibly high dimensional data. Dimension reduction for visualization and data analysis. Clustering, association rules, model fitting via the EM algorithm.	Sky Sheng	Varada Kolhatkar
DSCI 574	5	Spatial and Temporal Models	Time series. State space and change point detection. Hidden Markov Models. Gaussian processes.	Model fitting and prediction in the presence of correlation due to temporal and/or spatial association. ARIMA models.	Prajeet Bajpai	Prajeet Bajpai
DSCI 532	6	Data Visualization II	Interactive visualization, design choices, dynamic change over time, multiple views, data reduction, dealing with complexity.	How to make principled and effective choices with respect to marks, spatial arrangement, and colour. Analysis, design, and implementation of interactive figures. How to provide multiple views, deal with complexity, and make difficult decisions about data reduction.	Daniel Chen	Daniel Chen
DSCI 542	6	Communication and Argumentation	Claims, reasons, and evidence. Strengths and weaknesses of models. Effective oral and written presentation of scientific results, including interpretation of data and recognition of assumptions, bias, validity, and reliability. Citations, references, and peer-review.	How to interpret and present data science findings to a variety of audiences. Written and spoken presentation skills.	Grace Tompkins	Gittu George
DSCI 554	6	Experimentation and Causal Inference	Randomization. A/B testing. Blocked designs. Orthogonality. Batch effects, confounding. Causality. Contemporary examples. Simulations.	Statistical evidence from randomized experiments versus observational studies. Applications of randomization, e.g., A/B testing for website optimization. Methods for dealing with the multiple testing problem.	Ilya Musabirov	Ilya Musabirov
DSCI 575	6	Advanced Machine Learning	Neural networks trained with backpropagation. Deep learning. Overfitting and underfitting. Active data acquisition. Hyperparameter optimization.	Advanced machine learning methods in the context of natural language processing (NLP) applications. Bag of words, recommender systems, topic models, natural language as sequence data, Markov chains, and recurrent neural networks.	Elham E Khoda	Varada Kolhatkar
DSCI 591	7	Capstone Project	A capstone design project designed to give students experience in leading complex multidisciplinary projects relevant to data science.	A mentored group project based on real data and questions from a partner within or outside the university. Students will formulate questions and design and execute a suitable analysis plan. The group will work collaboratively to produce a reproducible analysis pipeline, project report, presentation and possibly other products, such as a dashboard.	MDS teaching team	MDS teaching team

MDS Event Calendar

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Event/Workshop	Date
Orientation (including Alumni Talks)	Aug 26th - 28th, 2026
MDS Dinner (including Alumni Talk)	Aug 28th, 2026
Semester 1 (Blocks 1-3) Midterm Break	Nov 9th - 13th, 2026
Mentoring Kick-off Event (for students registered in mentoring program)	Nov 2026
Holiday Party (MDS-V)	Dec 2026
Last Day of Semester 1 (Blocks 1-3)	Dec 21st, 2026
Start Day of Semester 2 (Blocks 4-6)	Jan 5th, 2027
Semester 2 (Blocks 4-6) Midterm Break	Feb 8th - 12th, 2027
Career Talk	Feb 2027
Capstone Fair	Feb 2027
Career Talk	Mar 2027
Career Talk	Apr 2027
End-of-program Celebration	End of June, 2027
Graduation Ceremony*	Late Nov, 2027
EDI Workshop	TBD
LinkedIn webinar	TBD
The Art of Building Connections for Job Search	TBD
Resume & Cover Letter Writing	TBD
ChatGPT for Jobseekers webinar	TBD
Interview Skills Workshop	TBD
Technical Interview Clinic	TBD
Employer Information session	TBD
Job Offer Negotiation	TBD
Post-Graduation Work Permit (PGWP) Workshop	TBD
Team Effectiveness Workshop	TBD
Career Panel	TBD
Mentoring Wrap-Up Event (for students registered in mentoring program)	TBD
Capstone Hackathon	TBD
Alumni Reunion	TBD

Note*: Please check https://graduation.ubc.ca/ for the date of graduation ceremony.

MDS Instructor 1 Position

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

The University of British Columbia, Vancouver invites applications for a tenure-track instructor-1 position, in the Department of Statistics. The position has a primary focus on contributing to the delivery and further development of the Master of Data Science (MDS) program, while also involving contributions to other departmental programs.
The MDS program is a collaborative effort of the Department of Computer Science, the Department of Statistics and the Faculty of Science.

This position provides an opportunity to pursue a career based on excellence and leadership in teaching, while participating in the intellectually exciting atmosphere of two top-tier departments jointly supporting the first comprehensive Masters of Data Science program in Canada.

MDS Policies

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Program Policies and Regulations

The Master of Data Science Policies and Regulations are part of the campus-wide UBC Policies and Regulations document. Below is a summary of those Policies and Regulations that students frequently ask about and that are specific to the Master of Data Science program.

Attendance

Attendance in lectures and lab sessions is expected. Your ability to succeed in the program depends on engaging fully with the material, your instructors, and your peers.

MDS Quiz Guidelines

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Overview

Unless otherwise specified or for project courses, each MDS course has two quizzes: one in week 3 and another in week 5. Students will have the option to attend a quiz review session to review their quiz results and submit regrade reqeusts. Please read through the quiz policy as well. The information in the quiz policy page has the final say in any discrepancies between the two pages.

Quiz and quiz review windows will appear on the MDS quizzes calendar with a ’tentative’ label until reservations open, at which point they are updated to ‘confirmed’. This is becasue quizzes are held at ORCA, a shared UBC testing facility, and we must schedule quiz availability around all other courses at UBC. We will do our best to keep the dates consistent throughout the program, provided the ORCA has adequate availability.

MDS Quiz Procedures

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Before your Quiz
- Reservations
- Cheatsheets
During your Quiz
After your Quiz
Alumni Cheatsheet Tips

Before Your Quiz

Making a Reservation

The following instructions apply to both quiz and quiz review reservations. You must make a separate reservation for each quiz, but only one quiz review reservation is required to review all quizzes from the previous week. Please make your reservations early, as availability is limited and time slots may fill up quickly. We cannot guarantee that your preferred time slot will be available.

MDS Quiz Procedures - CfA

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

What is the CfA?

The CfA stands for the Centre for Accessibility. It provides disability-related accommodations and programming designed to remove barriers for students with disabilities or ongoing medical conditions in all aspects of university life. For more information, visit the Centre for Accessibility.

If you have a disability or ongoing medical condition, you can request quiz accommodations through the CfA.

Accommodations

If you have a Letter of Accommodation (LOA) from the Center for Accessibility (CfA), to ensure that your approved accommodations are provided throughout the program, please:

MDS Tools

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Tool	Purpose	How to Access
Canvas	Gradebook	Log in using your CWL
GitHub Enterprise	Course materials and assignment distribution	Log in using your CWL
GitHub.com	Assignments for specific courses	Create your own account
Google Calendar	MDS schedule and deadlines	Access granted to the provided email address (usually a Gmail address)
Gradescope	Assignment submission	Access granted through Canvas*
PrairieLearn	Pre-lecture quizzes and MDS quizzes	Log in using your CWL
PrairieTest	Quiz and review session reservations	Log in using your CWL, invitation sent once per term
Slack	Primary communication tool	Invited via email
Workday	Final grades	Log in using your CWL

*Note: Gradescope username is the same as the Canvas primary email. Do not change your primary Canvas email address during the MDS program.

Mentor Information

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

If you are a Data Science professional who is interested in sharing your experiences, insights and advice with MDS students, please consider becoming a mentor. Mentors will also have the opportunity to network with other mentors and professionals during the events organized as part of the program.

How to Become a Mentor

Step 1: Read the program commitments and structure section and policies section below.

Mentoring Program

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Description

Since 2020, the UBC Master of Data Science (MDS) Mentoring program have welcomed Data Science professionals to mentor their students. Data Science professionals provide invaluable guidance and advice to MDS students, who are looking to build a career in the Data Science field upon graduation.

Important Pages

For more information on becoming a mentor
For more information on becoming a student mentee
For important dates and deadlines
If you have questions, please contact one of the MDS Mentoring Coordinators

Our Academic Team

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Courses in the UBC Master of Data Science - Vancouver program are primarily taught by our core teaching team. The team is in constant communication to share ideas, support each other, and collaboratively build the best program we can. These are the faces that you will be seeing every day throughout the MDS program.

Team Stats

13 : The total number of different languages we speak (including Cantonese, Farsi, Hindi, Japanese, Malayalam, Mandarin, Marathi, Nepali, Spanish, Swedish, and Vietnamese)
20 : The total number of coding languages we have experience working with as a whole.
10 : The number of different countries our team cheers for during the Olympics.
5.7 : The average number of years our team has of teaching experience (with a standard deviation of 3.2).
10 : The median number of postsecondary education years our team has studied.

Team Facts

The team often uses Bitmojis to communicate with one another.
The team is an assortment of morning birds (early risers) and night owls (late sleepers).
We have a teamwork contract that promotes a respectful and transparent work environment.
We hold weekly team meetings where we discuss not only our work progress but also our social wellbeing (this was implemented during COVID when we were working from home and we decided to keep it).

The Team

Prajeet Bajpai, Postdoctoral Research and Teaching Fellow

Prajeet received his Master's and PhD in Mathematics at the University of British Columbia and his Bachelor's degree at Dartmouth College. As a PhD student, he had the opportunity to work as a TA for MDS, and joined the program full-time as a Postdoctoral Teaching and Learning Fellow in 2024.

Prajeet likes to play tennis (and occasionally the guitar) and loves to cook and eat.

Learn more about Prajeet here.

Partner Info

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Benefits to Capstone Partners

You’ll have ~4 data scientists-in-training working on your project for ~2 months, at no cost. These students will be well versed in modern data science tools and techniques, including statistical analysis and visualizations using R and Python.
Students will create a data science product for your project (e.g., a dashboard, an analytic report, a set of scripts, a pipeline, or similar). This product is expected to be of high quality, with excellent code documentation and testing.

Past Projects

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Click the year to expand or collapse the projects that year.

2026:

AgriGates: Machine Learning Pipelines for Wearable IMU-Based Behavior Classification (Auto-Coding)
ALS GeoAnalytics: Rethinking Recruitment: Let LLMs Find the Right Fit
Applied AI Team, Klick Health: From Posts to Personas: Data-Driven Patient Segmentation from Social Media
BC Centre for Disease Control: Uncovering Opioid Poisoning and Dependence Patterns in Emergency Care Data
BC Children’s Hospital: Uncovering the Impact of Computerized Ordering on Pediatric Laboratory Testing
BC Children’s Hospital Research Institute, University of British Columbia: Reproducible Framework for Multi-Omics Analysis in Inflammatory Bowel Disease (IBD)
BC Parks: Image analysis of park infrastructure
Beaty Biodiversity Museum: Reconciling Data Authorities
Boreal Avian Modelling Centre, Biodiversity Pathways: Boreal Bird Dashboard: Model Insights for Improved Conservation
Clarius Mobile Health: Scanning the Landscape: Predictive Analytics of External Factors on Clarius Scanner Sales
Department of Fisheries and Oceans: From Data to Decisions: Automating Weekly Barkley Sound Salmon Stock Assessment Bulletins
Department of Fisheries and Oceans: Streamlining the West Coast of Vancouver Island Chinook salmon run reconstruction
Department of Statistics, University of British Columbia: How data scientists use LLMs
ETS: Equity by Design: Smarter Visuals for Smarter Decisions
GOcxm: Retail Display Compliance Automation Model
GOcxm: Image Forensics & Fraud Detection AI
IOTO International Inc.: Goverlytics
Justice Institute of British Columbia: Modeling costs of JIBC tuition programs
Kai Analytics International Inc. and Accrediting Commission for Community and Junior Colleges: Beyond the Balance Sheet: Linking Student Outcomes to Financial Sustainability
LiteFarm – University of British Columbia (UBC): Integrating Advanced Sustainability Analytics Into the LiteFarm Dashboard
PavePal Technologies Inc.: Evaluating Vision-Language Models for PASER-Based Asphalt Pavement Assessment from Imagery
PavePal Technologies Inc.: Grounded AI for Road Maintenance Decisions: Evaluating Retrieval Quality and Failure Modes
Providence Health Care - Foundry BC: Building Foundry Youth Journey Map
ResponsiveAds: Data-Driven Generation of Digital Ad Creatives
Seaspan Corporation: Automation of data extraction from sustainability documents for regulatory compliance
Square One Insurance Services: Predicting Home Attributes from Satellite Imagery
Square One Insurance Services: Content2Conversion - linking content themes to sales impact
Stellar Education: AI Student Progress Tracker
Tandem Research: Adverse patient outcome prediction with multimodal data
TeejLab: Machine Learning Classification of APIs in Code Repositories
Trilemma Foundation: Delivering Elite European Football (Soccer) Analytics
Trilemma Foundation: Improving Institutional Bitcoin Accumulation Strategies
UBC Faculty of Land and Food Systems - Animal Welfare Program: MooVision: Automated Detection of Cross-Sucking in Dairy Calves Using Computer Vision
Vaccine Evaluation Center, BC Children’s Hospital Research Institute: Data Integration of 35 years of vaccine preventable disease surveillance with IMPACT
Vaccine Evaluation Center, BC Children’s Hospital Research Institute: Medical Dictionary for Regulatory Activities [MedDRA] Autocoder
Vancouver Whitecaps Football Club: ChatVWFC – Player Search for Scouting
Vrify: Deep Learning–Based Prediction of Soil Geochemistry Using Multi-Source Raster Data

2025:

AgriGates: Advancing and Exploring Video Annotation and Auto-Annotation Models for Animal Behavior Studies.
ALS GeoAnalytics: A Framework for Carbon Emissions in AI Models
ALS GeoAnalytics: Explainable AI for Obtaining Mineralogical Insights: Visualizing Predictions in Deep Learning Models
Autograph Analytics: Online-to-Offline Attribution with Vehicle Sales and Digital Marketing
Bayes Studio Inc.: AI-Driven Wildfire Prediction and Data Integration Platform
Bayes Sutdio Inc.: AutoML CI/CD/CT: Continuous Training and Deployment Pipeline
BC Children’s Hospital - Digital Lab: Automated Medical Image Segmentation for Precision Orthopedic Surgery
BC Injury Research and Prevention Unit: Creating online interactive injury data visualizations
Brilliant Automation: Predictive Maintenance of manufacturing machinery
Canucks Sports and Entertainment: Analyzing Ticket Buyer Trends to Identify Potential Season Ticket Member
Creative Destruction Lab Vancouver: From Application to Graduation: Uncovering Patterns of Venture Success
Fathom: Dialogue2Data (D2D): Transforming Interviews into Structured Data for Analysis
FINLY TECHNOLOGY CORP.: Find me the better product!
HCL Technologies: AI-Based Visual Guidance System for the Visually Impaired. Image and Video Analysis Model for Safe Navigation in Building Premises
Heron Law Offices and AIMICI [civil society organization] (joint venture): Increasing Public Data Transparency for Immigration Law in Canada
heymate! Rewards Inc.: Customer retention supercharged.
LiteFarm project at the Centre for Sustainable Food Systems, UBC: Interactive Dashboard for Real-Time Agricultural Insights
Natural Resources Canada - Canadian Forest Service: Remote Sensing for Forest Recovery
Nautical Crime Investigation Services: Semi-supervised vessel trajectory analysis for unregulated fishing activity detection
Pacific Salmon Foundation: Pinniped Monitoring Machine Learning and Data Visualization Tools
Population Health Surveillance and Epidemiology Branch, Office of the Provincial Health Officer, Province of British Columbia: Forecasting excess mortality due to extreme temperatures in British Columbia Canada
Sensorimotor Physiology Laboratory: Is it me or is there someone in my head?
Sumit Kumar: Evaluating Large Language Models for Incident Resolution
The Valuation Standard -ViRA360: Quant AI-Powered IP Valuation
Trilemma Capital: Forecasting Bitcoin Transaction Fees
UBC Cybersecurity: Sniffing for Phishing
University of British Columbia: Diverse Data Repository for Data Science Education
University of Helsinki: Development of a Machine Learning Tool to Identify Biochemical Features of Proteins
Vancouver Whitecaps FC: Solving Set Pieces

2024:

ALS GoldSpot Discoveries Ltd.: Towards Decentralized Training of Machine Learning Models for Bone Cancer Detection
Analytika: Production Line Robotics Arms Vision Recognition
Autozen Technology: Autozen Recommendation Engine
Boeing Digital Solutions, Inc. d/b/a Jeppesen: Navigation Chart Change Detection
Canucks Sports & Entertainment: Ticket Pricing Analysis of the Vancouver Canucks
Changing Ocean Research Unit, Institute for the Oceans and Fisheries, UBC: Can you afford sustainable seafood in the future under climate change?
Faber Connect: Predicting No-Shows and Job Success in Construction
Illuminex AI: Airfield hazards: bird tracking at airports
Impact Canada (Impact and Innovation Unit, Privy Council Office): Measuring innovation through knowledge production
Litefarm / University of British Columbia: Leveraging data science for agricultural carbon and biodiversity outcomes
Rio Tinto Exploration: FaultSENS
Rio Tinto Exploration: Project DrillSense
Rocketbrew: Generative AI Icebreaker - Wildcard Clickbait 🃏
SRK Consulting (Canada) Inc.: Benchmarking tailings facility risks
SRK Consulting (Canada) Inc.: Developing novel methods for liquefaction prediction
Seahorse Strategies: UBC Stock Portfolio Allocation Formula
Side.: AI-Driven Real Estate Insights: Revolutionizing Pre-Construction Sales
Strait of Georgia Data Center: Survival Analysis System for Salmon in the Salish Sea
TRIUMF and UBC: AI CALRICH
University of British Columbia: Checklists and LLM prompts for efficient and effective test creation in data analysis
Vancouver Whitecaps Football Club: The Global Game: Ranking Soccer Clubs Worldwide
Vancouver Whitecaps Football Club: Predicting Physical Performance of Football Players
Westland Insurance: Augmenting Customer Retention Modelling with NLP Feature Engineering

2023:

Autozen Technology: Autozen Valuation Guru
BGC Engineering Inc.: Predicting the largest floods in Canadian rivers
Citysage for Reliance Foundry: Noise Pollution: Spatial modelling and visualization of urban sound levels
Cymax Group Technologies: Product Knowledge Graph
EOAS department, UBC Faculty of Science: Data Science for polar ice core climate reconstructions
ALS GoldSpot Discoveries Ltd.: A Study on the Effects of Representation Bias on AI Performance and Methods of Mitigating it
Insurance Corporation of BC: Image Recognition of Vehicle Odometer Readings
Neotoma Paleoecology Database(University of Wisconsin – Madison): Finding Fossils in the Literature
Olyns: The Price Is Right!
Rio Tinto Exploration: Clean sat
Seaspan Corporation: Development of operation and maintenance analytics platform for container ships
Sitewise Analytics: Restaurant Segmentation Analysis
Slalom Consulting: Power Price Prediction - a short-term forecast
TRIUMF and UBC: CALORICH AI
Trusting Pixels Inc.: Compressed Softening Filter Detection
UBC Digital Emergency Medicine: Predictive analytics to support HLBC 8-1-1 and HEiDi triage
Vancouver Whitecaps Football Club: Finding Football Talent with Wearable Technology Using PlayerMaker sensors to understand academy player performance
Vancouver Whitecaps Football Club: Terrific Touch
Westland Insurance: Predicting Customer Conversion
aDolus Inc: Can AI spot risky software in critical infrastructue?

2022:

Canadian Campus Wellbeing Survey/UBC: Impact of COVID-19 on student mental health: Lessons from the CCWS
Corporate Finance Institute Education Inc.: Recency, Frequency, and Monetary Value Analysis
Corporate Finance Institute Education Inc.: Sales Forecasting
Defined Finance Ltd.: DeFi Dashboard: Follow and Forecast the Money
Glentel: Practical people analytics for predicting retention
Glentel: Practical people analytics for predicting employee performance
ALS GoldSpot Discoveries Ltd.: Detection and Mitigation of Data Drift and Model Decay
ALS GoldSpot Discoveries Ltd.: Panorama stitching of core-photos
Olyns: Prune CNN models to help people go green
OraQ AI: Using NLP to untangle the complex web of dental conditions
Population Health Surveillance and Epidemiology: BC Chronic Disease Visualization and Trend Analysis with R Shiny
Reliance Foundry Co. Ltd.: LiDAR object detection and classification for cities
Riversol Skincare Solutions Inc: Forecasting the success of online lead generation
Rocketbrew Inc.: Creating (figurative) ecommerce shopping aisles with ML 🛒🛒🛒
Seahorse Strategies: Data Analytics for Stock Market Trading
Sensing in Biomechanical Processes Lab (SimPL): Towards a simplified method for video confirmation of head impact events in contact sports
Sensorimotor Physiology Laboratory: Decomposition of muscle activity for sensorimotor neuroscience
Sitewise Analytics: Determining Restaurant Sales Performance Drivers through Feature Selection
Suncor Energy Inc.: Modelling Heat Exchanger Units to Optimize Cleaning Schedules
TRIUMF: RICH AI
Trusting Pixels Inc.: IMAGE COMPARISON ANALYSIS
Trusting Pixels Inc.: PHOTO WITHIN PHOTO DETECTION
UBC Animal Welfare Program: Cow bonds: Visualizing and assessing changes in the social networks of dairy cows
UBC Sauder and TeejLab: An Analytical Framework for Quantifying API Risks
Vancouver Whitecaps Football Club: Passing Perfection - Using Optical Tracking and Event Data to Evaluate MLS Player’s Passing Tendencies
Vancouver Whitecaps Football Club: Disruptive Defending - Using Optical Tracking and Event Data to Evaluate MLS Players’ Defensive Performance
Viewpoint AI: Life Decision Support: Choose your best career path
Westland Insurance: Predicting Customer Retention
World Bank: How quickly can South Asia transition to a green economy?

2021:

Analytika: Transforming Customer Experiences
BC Centre on Substance Use: Using data science to identify and visualize novel compounds in illicit drug checking samples
BC Stats: Understanding voting method choices in the 2020 BC General Election
British Columbia Investment Management Corporation (BCI): What Can SEC 10-K Textual Disclosures Tell Us About a Firm’s Earnings Quality and Future Stock Returns?
Canada Web Analytics Team: Determining the Use Cases Across Data Science Sub-Fields for the Government of Canada's Web Analytics Operations
Cenovus: Using Time Series Temperature Data to Determine Well Productivity
Collaboration for Outcomes Research and Evaluation (CORE): Data science and health outcomes research
Dialpad: Detecting Emerging Topics, Trends and Anomalies from Call Center Transcripts
Dialpad: Understanding & Predicting Customer Satisfaction Using Vocal Features
Gerontology and Diabetes Research Laboratory (GDRL): Machine Learning Approaches to: 1. Diagnosing Lipohypertrophy at the bedside, and 2. Falls Prediction in Long Term Care
Glentel: People Analytics
ALS GoldSpot Discoveries Ltd.: Automated drill core logging through the lens of Machine learning and Deep learning
Hootsuite: Customer Segmentation using Hootsuite Product Usage Data
Newsly: Audio listening preferences
Orbis Investments: Earning Calls Deception Analysis
PayByPhone: Anomaly Detection
Properly Inc: Image Processing: Quantifying The Home Condition From Property Images
Realtor.com: Identifying real estate investment opportunities using Machine Learning
Realtor.com: Will they or won't they? Return user prediction
Sensing in Biomechanical Processes Lab (SimPL): Extracting and visualizing the human brain state using EEG data
UBC Cybersecurity Group: Defend UBC
Vancouver Whitecaps Football Club: Football Fortune Telling: Predicting MLS Performance
Vancouver Whitecaps Football Club: Match Fit – Using Optical Tracking Data to Evaluate MLS Players’ Power, Fitness & Fatigue
aDolus Inc: Software File Clustering (What is this file?)

2020:

aDolus: Unearthing Hidden Vulnerabilities in Mission Critical Software
Analytika: Smart Agriculture
Analytika: Wells Timelines
BC Stats: Text Analytics: Quantifying the Responses to Open-Ended Survey Questions
BGC Engineering: Automated Tailings Dam Detection from Satellite Data
BGC Engineering: Data Driven Flood Forecasting
Biba Ventures: Using Machine Learning to Predict Playground Usage Across the Continent
The Deetken Group: Forecasting the Evolution of Vancouver's Business Landscape
E-Comm 911: Natural language processing to help save lives and protect property
Glentel: Making sense of people data
ALS GoldSpot Discoveries Ltd.: Core Photo Analysis
MDA: Image Captioning of Overhead Earth Observation Imagery
Merchant Growth: Merchant Score: Intelligent Credit Decisioning For Risk Management
PayByPhone: Crowdsourced parking locations
Realtor.com: Photo-realistic Neighborhood Image Synthesis
Riversol Skincare Solutions: E-commerce domination in highly competitive markets driven by data science
Seahorse Strategies: Seahorse Momentum Indicator
TransLink: Vision over Transit Incidents & Claims
TransLink: Understanding Bus Delay in Metro Vancouver
TransLink: Optimizing Transit Stops
UrbanLogiq: Analysis of Connected Vehicle Driving Behaviour as a Predictor of Accidents
UrbanLogiq: Contextual analysis of amenity gaps in at-risk communities
Vancouver Whitecaps Football Club: Modelling the Physical Performances of the Vancouver Whitecaps
Vancouver Whitecaps Football Club: Understanding Players' Offensive and Defensive Performance in Major League Soccer

2019:

BC Stats: Quantifying the Responses to Open-Ended Survey Questions
BGC Engineering: Automated Landslide Detection and Delineation from Digital Terrain Data
British Columbia Maritime Employers Association: Improving Labour Forecasting to Promote the Competitiveness of BC Ports
E-Comm 911: Predictive Staffing Model to Help Save Life and Protect Property
Fresh Prep: Forecasting Meal Kit Orders
Management Information Systems Group, UBC Sauder School of Business: Extracting a Corporate Social Network from SEC Filings
Minerai de fer Québec / Quebec Iron Ore: Image recognition of rock types for identification of rock formations
Minerai de fer Québec / Quebec Iron Ore: Predicting geological properties from drill metrics to predict rock composition
Providence Health Care: Forecasting of Staffing Needs
QxMD: Generate cross-product recommendations to help get medical research adopted in clinical practice
QxMD: Match real-time news stories with medical research literature
RStudio: What the Git Is Going On Here!?
Realtor.com: Estimate the Value of Key Local attributes used in buying decisions
Seahorse Strategies: Predicting the Stock Market
TRIUMF: π-e-μ AI
TeejLab: Technical Legal Risk Assessment for Data Services
Tetrad: Understanding Restaurant Sales
UrbanLogiq: Indicators of Crash Severity

2018:

BC Stats: Discovering thematic categories from survey comments
BGC Engineering: Anomaly detection and flood forecasting using real-time hydrometric data
Destination BC: Predicting conversion rates for tourism advertisements on Facebook and Instagram
Finn AI: Evaluating a Natural Language Processing Pipeline for Chatbots
Microsoft MSN: Web traffic prediction for msn.com
SAP and Teejlab: Automated Legal Risk Assessment on Web Service License Changes
Semios: Binary Classification of Leaf Wetness Using Sensor Data
Thinkific: Success in online learning: recommending actions to course creators
Unbounce: Using survival analysis to finding leading indicators of customer churn
QxMD: Building a Recommendation System for Medical Research Papers
UBC Sauder School of Business: Extracting features from financial documents for predicting firm performance
Visier: Automated Human Resources Insight Discovery

2017:

BC Stats: Empowering employee engagement through AI
Coast Mountain Bus Company: Forecasting Transit Schedules and Congestion Areas
Rio Tinto: Tools for Analyzing Mining Drill Data
UBC CTLT: edXvis: Interactive Visualization of Student Engagement with edX MOOCs
Unbounce: Unbounce Community Forum Analysis

Proposals

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Capstone Project Proposals open in late summer/early fall each year, see the timeline for this year’s submission deadlines.

Partners may choose to submit a capstone project(s) to the MDS Vancouver program (general data science) and/or the MDS Computational Linguistics program (language-related data science). If you’re unsure about which program to submit your project to, read more below.

Proposal forms can be found here:

Should I submit my proposal to MDS Vancouver or MDS Computational Linguistics?

The MDS Vancouver (MDS-V) program covers all aspects of data science, including topics of data wrangling, visualisation, dashboards, statistics and machine learning, amongst others. You can read more about the program here and can see the capstone page on this website to learn more about the type of projects MDS-V addresses in capstone.

Resources

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

For prospective or entering students

MDS self-test for prospective and entering students.
Answers for the above self-test.
Online courses and other resources to help prepare for the MDS program.
Advice for applying to the MDS program

For current MDS students

General policies and guidelines

Labs and quizzes

Student rep and feedback

Resources

Possibly of general interest

Sample Capstone Proposal

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Organization Name

Management Information Systems Group, UBC Sauder School of Business

About Your Organization

We are faculty members at UBC Sauder School of Business. Big Data and Data Science are creating profound impacts on various fields including Information Systems research. Our research interests include Business Analytics, which has the aim of gaining business insights from internal and external data sources in order to make timely data-driven decisions for competitive advantage in the complex business environments.

Slack in MDS

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Slack

We will be using Slack as our primary means of electronic communication throughout the MDS program. We will invite you to our UBC-MDS Slack workspace shortly before the program starts. Please note that you will be removed from this workspace around mid-July, after you have completed the program.

At some point, we will also invite you to the UBC MDS Alumni Slack workspace, which is completely separate from the main UBC-MDS workspace you use as a student. You will not lose access to this workspace after graduation and can stay connected with other MDS alumni.

Student feedback

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

The MDS team received feedback from the MDS students in two primary ways: monthly feedback sessions with the block rep, and the formal UBC teaching evaluations. These document attempts to clarify the roles of the these two feedback channels.

We see the roles of block rep feedback vs. the official UBC course evaluations as roughly formative vs. summative assessment, respectively. That is, the block rep’s role is to give us information to help us improve, rather than give us an A+ or a C-. These results will be heard/read by the core MDS team, and the information will be used to improve the program. Since we get this feedback during the block, we can sometimes make immediate changes within a given course. The formal UBC course evaluations, on the other hand, come after the course ends. They are also seen by a different group of people, including the department heads and the Dean. These will become part of the instructors’ “files” and may influence things like tenure/promotion, or raise a red flag if the evaluations are very poor. While we certainly do read all the comments on the official evaluations (and therefore they are also useful in improving our teaching), you can also think of them as a place to record your thoughts on our teaching quality. Whether or not these evaluations actually do a good job of evaluating teaching is a topic of heated debate, though; see e.g., this article.

Student Info

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

The goal of MDS Capstone Project is to provide technical and professional Data Science training and to reinforce the skills you learned in the course work in a realistic situation. It will provide you with hands-on experience on a current and relevant data science project to tackle real problems faced by our partners. We aim to make this experience as close as possible to a professional experience for the students. Naturally, this project has a pedagogical component, but students can expect to have an engaged and interested partner to support them.

Student Information

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

The UBC Master of Data Science Mentoring Program gives students the opportunity to connect with a Data Science professional who shares their experience in the field of data science. You must be a current UBC MDS student to participate in this program.

How to Become a Student Mentee

Step 1: Read the program commitments and structure section and policies section below.

Student Representatives

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

MDS Student Representatives

In the new academic year, there will be one set of student representatives in Term 1 (Blocks 1-3) and another set in Term 2 (Blocks 4-6). Additionally, there will be a third set of student reps during the Capstone project period. Student representatives volunteer to represent their peers, acting as a liaison between students and MDS leadership. This allows us to monitor how the program is progressing and make adjustments if needed.

Timeline

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

This year’s capstone course will run April 26 - June 24, 2027 (~8 weeks).

The soft deadline for UBC Capstone proposals is Oct 31st, 2026. And the hard deadline is Nov 30th, 2026. Proposals sent in by Oct 31st will receive feedback and prospective capstone partners will be offered an opportunity to revise their proposal (which will hopefully increase their chances of being selected).

Event	Dates
Call for Capstone proposals opens	August 1, 2026
Capstone Info Session	September 18, 2026
Deadline to submit proposals for early feedback	October 31, 2026
Final deadline to submit proposals	November 30, 2026
Selection round one: staff	January 5, 2027
Capstone fair	February 5, 2027
Selection round two: students	February 7, 2027
Partners demonstrate readiness of data and sign legal docs	March 1, 2027
Students are assigned to projects	March 31, 2027
Kickoff meetings	April, 2027
Kickoff hackathon	April 26 - April 28, 2027
Project	April 26 - June 24, 2027

Proposal revision process

Members of an organization interested in participating as a capstone partner should submit a project proposal.

UBC MDS Vision

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Mission Statement - short version

Our purpose is to teach and advocate for the responsible use of data science.

Mission Statement - long version

Our purpose is to teach and advocate for the responsible use of data science. We define data science as the process of gaining insight from data through reproducible and auditable processes. Responsible use of data science involves insisting on reproducible practices, carefully considering the ethical implications of one’s work, clearly communicating and not overstating one’s results, creating effective analyses and visualizations, as well as minimizing technical debt. We are committed to teaching these practices to our students, and disseminating them within the global data science community both directly and indirectly through our alumni. Finally, we are committed to be involved in determining the definition, scope, and boundaries of data science.

UBC resources

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Study spaces on campus

Within the Computer Science building (ICICS)
- ICCS 108 (Bullpen, the computer area attached to ICCS 144 Graduate Student Lounge)
  - ICCS 108 Guidelines
- ICCS 359 (MDS Lounge)
- ICCS X860 (Department Lounge)
  - Please note this room may be booked for events.
- Book a meeting room
List of UBC libraries and hours
- If you prefer to book a private room to study, check bookable study spaces.
Informal learning spaces

Computing resources

The UBC Library provides access across the university to print and electronic books, journals and more.
The ICICS/Computer Science Reading Room is an independent departmental reading room located in ICCS 262. Our area of specialization is information technology and you are welcome to visit during opening hours.
You may contact the course coordinators to request a loaner laptop for a few days if your personal computer is temporarily unavailable. Students are still expected to have their own laptop that meets the minimum requirements.

Printers

Computer Science building (ICICS): Printer Locations
Complete list of public printers in the libraries
- Note: payment required

Student services

The About UBC Student Services page provides a comprehensive directory of resources available to students, including academic support, health and wellbeing, community services, and more.
International students should checkout the International Student Guide.

Other resources

Ubuntu

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Installation notes
UBC Student Email
Ubuntu software settings
Web browser
Password manager
Slack
Visual Studio Code
GitHub
Git
Python, Conda, and JupyterLab
R, IRkernel, and RStudio
Quarto CLI
LaTeX
PostgreSQL
Docker
VS Code extensions
Improving the bash configuration
Post-installation notes

Installation notes

If you have already installed Git, Latex, or any of the R or Python related packages please uninstall these and follow the instructions below to reinstall them (make sure to also remove any user configuration files and backup them if desired). In order to be able to support you effectively and minimize setup issues and software conflicts, we require all students to install the software stack the same way.

Useful Textbooks

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

About this document

This document is a compilation of textbooks referenced in MDS courses. The original version was created by an MDS student, Talha Siddiqui.

Textbook	Author(s)	Year	Course Code	Course Name	Comments
A Course in Machine Learning	Hal Daumé III	2017	DSCI 571, 572, 573, 563, 575	Supervised Learning I, Supervised Learning II, Feature and Model Selection, Supervised Learning II, Advanced Machine Learning
Advanced R	Hadley Wickham	2014	DSCI 511	Programming for Data Science	This is a prominent resource for R as a programming language, allowing the reader to dig deep into R. It anticipates readers to already have some programming background. Its first part on Foundations is closely aligned with the objectives of DSCI 511, and is therefore the textbook for the second half of the course. Gaining familiarity with this book will likely be an asset in your data science career.
Algorithm Design	John Kleinberg and Eva Tardos	2005	DSCI 512	Algorithms and Data Structures
Algorithm Design: Foundations, Analysis, and Internet Examples	Michael Goodrich and Roberto Tamassia	2001	DSCI 512	Algorithms and Data Structures
Algorithms	Sanjoy Dasgupta, Christos Papadimitriou and Umesh Vazirani	2006	DSCI 512	Algorithms and Data Structures
An Introduction to Statistical Learning: with Applications in R	James, Gareth; Witten, Daniela; Hastie, Trevor; and Tibshirani, Robert	2014	DSCI 561, 563, 572, 573	Regression I, Unsupervised Learning, Supervised Learning II, Feature and Model Selection	For 561: Especially Chapter 3, A modern and approachable take on statistics / machine learning. For 573: Chapter 2: Statistical Learning, Chapter 5: Resampling Methods, Chapter 6: Linear Model Selection and Regularization, Chapter 7: Moving Beyond Linearity
Art of Data Science	Roger Peng & Elizabeth Matsui	2016	DSCI 522	Data Science Workflows
Artificial intelligence: A Modern Approach, 3rd Edition	Russell, Stuart and Peter Norvig	2009	DSCI 571, 572	Supervised Learning I, Supervised Learning II
Artificial Intelligence: Foundations of Computational Agents, second edition	David Poole and Alan Mackworth	2017	DSCI 571, 572	Supervised Learning I, Supervised Learning II
Bayesian Data Analysis	Andrew Gelman, John Carlin, Hal Stern, David Dunson, Aki Vehtari, and Donald Rubin		DSCI 553	Statistical Inference and Computation II
Data Analysis and Visualization Using R	David Robinson	2014	DSCI 511	Programming for Data Science
Data Wrangling with Python: Tips and Tools to Make Your Life Easier	Jacqueline Kazil, Katharine Jarmul	2016	DSCI 523	Data Wrangling
Database Management Systems, 3rd Edition	Ramakrishnan, Raghu and Gehrke, Johannes	1996	DSCI 513	Databases and Data Retrieval
Deep Learning	Ian Goodfellow and Yoshua Bengio and Aaron Courville	2016	DSCI 572	Supervised Learning II
Deep Learning With Python	Jason Brownlee		DSCI 572	Supervised Learning II
Dive into Deep Learning	Aston Zhang, Zack C. Lipton, Mu Li, Alex J. Smola		DSCI 572	Supervised Learning II
Extending the Linear Model with R: Generalized Linear, Mixed Effects and Nonparametric Regression Models, Second Edition	Julian J. Faraway	2016	DSCI 562	Regression II
ggplot2 Elegant Graphics for Data Analysis	Hadley Wickham	2009	DSCI 531	Data Visualization I	Readable, comprehensive resource for learning about ggplot2, by the main author of the ggplot2 package, Hadley Wickham.
Grokking Deep Learning	Andrew Trask	2019	DSCI 572	Supervised Learning II
Hands-On Programming with R	Garrett Grolemund	2014	DSCI 511	Programming for Data Science
Houston, We Have a Narrative: Why Science Needs Story	Randy Olson	2015	DSCI 542	Communication and Argumentation	Writing & Speaking
Information Theory, Pattern Recognition and Neural Networks	David J.C. MacKay	2003	DSCI 563	Unsupervised Learning	Chapters 20-22
Introduction to Algorithms, 3rd edition	Thomas H. Cormen, Charles E. Leiserson, Ronald L. Rivest and Clifford Stein		DSCI 512	Algorithms and Data Structures
Introduction to Data Mining	Pang-Ning Tan, Michael Steinbach, Vipin Kumar	2005	DSCI 572	Supervised Learning II
Introduction to Empirical Bayes: Examples from Baseball Statistics	David Robinson		DSCI 553	Statistical Inference and Computation II
Introduction to Machine Learning with Python: A Guide for Data Scientists	Andreas C. Mueller and Sarah Guido	2016	DSCI 571	Supervised Learning I
Introductory Time Series with R	Cowpertwait, P. and Metcalfe, A.	2009	DSCI 574	Spatial and Temporal Models	A great hands-on approach to time series modelling
Linear Models with R	Julian James Faraway	2005	DSCI 561	Regression I	Comprehensive book on linear models.
Machine Learning: A Probabilistic Perspective	Kevin Murphy	2012	DSCI 572	Supervised Learning II
Mathematics for Machine Learning	Marc Peter Deisenroth, A Aldo Faisal, and Cheng Soon Ong	2018	DSCI 572	Supervised Learning II
Mining of Massive Datasets 2nd Edition	Jure Leskovec, Anand Rajaraman, Jeffrey David Ullman	2014	DSCI 572	Supervised Learning II
Modern Dive: An Introduction to Statistical and Data Sciences	Chester Ismay and Albert Y. Kim	2018	DSCI 552	Statistical Inference and Computation I
Neural Networks and Deep Learning	Michael A. Nielsen	2018	DSCI 572	Supervised Learning II
OpenIntro Statistics	David M Diez, Christopher D Barr, Mine C ̧etinkaya-Rundel	2010	DSCI 552, 561	Statistical Inference and Computation I, Regression I	Fairly accessible, seems to lean towards a traditional approach. Chapters 7 & 8 are relevant for linear regression
Pattern Recognition and Machine Learning	Christopher Bishop	2007	DSCI 572	Supervised Learning II
Probabilistic Programming and Bayesian Methods for Hackers	Cam Davidson-Pilon		DSCI 553	Statistical Inference and Computation II
Python Data Science Handbook	Jake VanderPlas	2016	DSCI 511	Programming for Data Science
Python for Computational Science and Engineering	Hans Fangohr	2016	DSCI 511	Programming for Data Science
Python for Data Analysis: Data Wrangling with Pandas, NumPy, and IPython	Wes McKinney	2011	DSCI 511	Programming for Data Science
R for Data Science (r4ds)	Garrett Grolemund & Hadley Wickham	2016	DSCI 531, 542, 561	Data Visualization I, Communication and Argumentation, Regression I	For 531: Overall good book on using R for data science – including data vis, of course! For 542: Tools & Technology Chapters 26-30. For 561: Especially Part IV, Practical and approachable book on the use of R for data science.
R Graphics Cookbook	Winston Chang	2012	DSCI 531	Data Visualization I	Good as a reference if you want to learn how to make a specific type of plot in ggplot2.
Spatio-Temporal Methods in Environmental Epidemiology	Shaddick, Gavin and Zidek, James V.	2016	DSCI 574	Spatial and Temporal Models	A less detailed treatment of time series analysis as it is not a primary focus of the book. A good addition in terms of examples to the lecture notes. Chapters 10.3, 10.4, 10.6
Statistical Rethinking: A Bayesian Course with Examples in R and Stan (& PyMC3 & brms too)	Richard McElreath		DSCI 553	Statistical Inference and Computation II
Survival analysis: a self-learning text, 3rd edition	David G. Kleinbaum, Mitchel Klein	2012	DSCI 562	Regression II	Non-technical explanation of survival analysis, with a nice succinct summary along the side of each page. Recommends epidemiological background, but we will avoid those parts.
The Analysis of Time Series: An Introduction	Chatfield, Chris	2003	DSCI 574	Spatial and Temporal Models	Chapters 1-5 A very readable introduction to time series analysis, without heavy mathematics
The Art of Computer Programming, Volume 1-4	Donald E. Knuth		DSCI 512	Algorithms and Data Structures
The Elements of Statistical Learning. Second Edition	Hastie, T., Tibshirani, R. and Friedman, J.	2009	DSCI 563	Unsupervised Learning
The Psychology of Persuasion	Robert Cialdini	1984	DSCI 542	Communication and Argumentation	Persuasion: Influence
The Sense of Style	Steven Pinker	2014	DSCI 542	Communication and Argumentation	Writing
Think Python: How to Think Like a Computer Scientist	Allen B. Downey	2002	DSCI 511	Programming for Data Science	Standard textbook for introductory programming courses. It includes case studies and exercises.
Thinking, Fast and Slow	Daniel Kahneman	2011	DSCI 542	Communication and Argumentation	Heuristics & Biases
Visualization Analysis and Design	Tamara Munzner	2014	DSCI 531, 532	Data Visualization I, Data Visualization II	The go-to book for data vis theory

Windows

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Installation notes
UBC Student Email
Web browser
Password manager
Slack
Visual Studio Code
GitHub
Git, Bash, and Windows Terminal
Python, Conda, and JupyterLab
R, IRkernel, Rtools, and RStudio
LaTeX
Make
PostgreSQL
Docker
Quarto CLI
VS Code extensions
Improving the bash configuration
Post-installation notes

Installation notes

If you have already installed Git, Latex, Make, or any of the R or Python related packages please uninstall these and follow the instructions below to reinstall them (make sure to also remove any user configuration files and backup them if desired). In order to be able to support you effectively and minimize setup issues and software conflicts, we require all students to install the software stack the same way.

With Great Data Science Comes Great Responsibility

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

Warning signs and tools to stop the spread of misinformation.

Have you ever seen an over-enthusiastic advertisement and wondered how much of it was true? If so, you’re already one step ahead of 67% of people! Okay… that was completely made up. But hopefully, you can see how easy it can be to trust numbers as facts, especially when they are embedded in nice compliments. With a world of information right at our fingertips, it has become increasingly difficult to decipher what’s true, what’s false, and what’s true with a little twist.

Working with Others

info-mds@science.ubc.ca — Mon, 01 Jan 0001 00:00:00 +0000

11 Tips on How to Work Well with Others. By Angela Pau.

1. Establish clear goals and ground rules

Set well defined project goals when working in a team. It is important that every member of the team is clear on what you would like to achieve as a team and how they will know if the team has succeeded. Also as a team it is important to set ground rules on how you are going to work together. What are the timelines of this project? How often will you meet? Which roles will each of you take? What is the best method to maintain communication?

UBC MDS

The Purpose of an Education Isn’t What You Think it Is

Doing homework and passing exams teach you more than just the course materials, but they also train you to be a motivated problem-solver.

N-of-1 Trials and Analyzing Your Own Fitness Data

My Little Markov Model - Now Tweeting New Taylor Swift Lyrics

Picture This…

MDS Capstone via Zoom: A lookback

Project Management

Have a realistic timeline, give a buffer for your intended timeline

Generating Photo-Realistic Neighbourhoods using Artificial Intelligence

NLP in the Real World: A Reflection on a Two-Month Capstone

MDS 2019/2020 Capstone Seminar Series

Integrating R & Python into a Data Science program

Project courses in MDS

What's for dinner? Predicting customer order probabilities

Winning the EasyMarkit AI Hackathon

About the EasyMarkit AI Hackathon

Teaching Convolutional Neural Networks

Designing a Master of Data Science program: goals, design decisions, and lessons learned

Stat-CS partnership

Welcome to our 2018-19 cohort

Bringing data science to new industries

Learning from real data

Becoming a data scientist: My year-long hiatus from medical school

Our curriculum, Part 1: Computer science & machine learning

Visualizing massive open online courses

Statistics-ML dictionary

Communication in data science: more than just the final report

Algorithms and optimization

Introducing our 2017-18 cohort

Teaching with GitHub

Postdoctoral Teaching & Learning Fellow

UBC MDS self-test

How to use this document

About UBC MDS

Application Tips

1. Understand the program

Capstone Projects

Career and Industry Resources

Career and Professional Development

Salary Information

Data Science

Vancouver Meetups

Facebook

Newsletters

Associations

Industry

Code of Conduct

MDS Leadership Code of Conduct

Our Pledge

Contributing to the MDS Blog

1. Reach Out

Data science terminology

About this document

Stat-ML dictionary

regression

bias

parameter and parametric

early stopping

hypothesis

factor

kernel

Dates and Deadlines

Contacts

DSCI 511

Short Description

Learning Outcomes

Reference Material

DSCI 512

Short Description

Course Learning Objectives

Reference Material

Instructor (2016-2017)

DSCI 513

Short Description

Course Learning Objectives

Reference Material

Instructor (2016-2017)

DSCI 522

Short Description

`regression`

`bias`

`parameter` and `parametric`

`early stopping`

`hypothesis`

`factor`

`kernel`