sentiment_analysis

Source: R/tweetr.R

This function takes a tweet dataframe as input. The input dataframe should contain a column named 'tweet' that contains tweet text information. The function cleans the text inn 'tweet' column by removing http component, punctuation, end words, reduce the letters to lowercase and word stemming. Then the function matches each word to the sentiment 'positive' or 'negative' using the tidytext 'Bing' lexicons. And output is a dataframe that contains words used in the tweet texts and assign each word with either 'positive' or 'negative' sentiment plus sorting by the numbers of appearence of that word.

sentiment_analysis(tweet)

Arguments

tweet

data.frame

Value

tweet_result data.frame

Examples

tweet_data <- tweetr::brunomars_tweet
sentiment_analysis(tweet_data)

#> Joining, by = "word"
#> Joining, by = "word"
#> # A tibble: 126 x 3
#>    word      sentiment     n
#>    <chr>     <chr>     <int>
#>  1 love      positive     40
#>  2 happy     positive     25
#>  3 awesome   positive      4
#>  4 killed    negative      4
#>  5 beautiful positive      3
#>  6 blow      negative      3
#>  7 cool      positive      3
#>  8 fuck      negative      3
#>  9 magic     positive      3
#> 10 merry     positive      3
#> # ... with 116 more rows