Visualizing sentiment by complaint type
Now let's create a sentiment_twitter data frame and see whether or not complaints really are more negative, on average.
Cet exercice fait partie du cours
Introduction to Text Analysis in R
Exercice interactif pratique
Essayez cet exercice en complétant cet exemple de code.
sentiment_twitter <- tidy_twitter %>%
# Append the bing sentiment dictionary
___(___) %>%
# Count by complaint label and sentiment
___(___) %>%
# Spread the sentiment and count columns
___(___) %>%
# Compute overall_sentiment = positive - negative
___(___)