Perl6 package for scoring the emotional content of a piece of text from its word use.
Currently the AFINN data set from Twitter usage is used to give base word scores for sentiment. https://github.com/fnielsen/afinn
AFINN is a list of English words rated for valence with an integer between minus five (negative) and plus five (positive). The words have been manually labeled by Finn Årup Nielsen in 2009-2011. The file is tab-separated.
'emobot' included is an example IRC bot using Text::Emotion::Scorer and Net::IRC::Bot.
#!/usr/bin/env perl6 use v6; #Create a Scorer object and all of the sentiment data is loaded for your use use Text::Emotion::Scorer; my $emotion = Text::Emotion::Scorer.new; #Score a whole passage of positive sounding text say $emotion.score("I really love Perl6. Hurrah!"); #Score a whole passage of negative sounding text say $emotion.score("I hate Monday's they suck!"); #Just get the word scores the passage score is working with say $emotion.score_word("failure");