search

trinker / sentimentr

Dictionary based sentiment analysis that considers valence shifters
Other
427 stars 84 forks source link

I wonder why this sentence is recognized as positive #127

Closed dogonanisland closed 3 years ago

dogonanisland commented 3 years ago

I wonder why this sentence is recognized as positive.

sentiment('I did not like how it was boring') element_id sentence_id word_count sentiment 1: 1 1 8 0.1767767

trinker commented 3 years ago

Beause the 'not' was also associated witht he 'boring' window. Here we take n.before from 5 down to 4:

sentiment('I did not like how it was boring', n.before = 4)
##    element_id sentence_id word_count  sentiment
## 1:          1           1          8 -0.5303301
dogonanisland commented 3 years ago

Thanks for getting back to me. I see the effect of n.before. I have a bunch of texts like these clause sentences (students' responses to an open-ended item on a survey). What is the best way to conduct the sentence-level sentiment analysis? Do I have to set up n.before (n.after for certain cases) each sentence at a time?

Thanks, Atsushi Miyaoka

On Fri, Oct 8, 2021 at 7:42 PM Tyler Rinker @.***> wrote:

Beause the 'not' was also associated witht he 'boring' window. Here we take n.before from 5 down to 4:

sentiment('I did not like how it was boring', n.before = 4)

element_id sentence_id word_count sentiment

1: 1 1 8 -0.5303301

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/trinker/sentimentr/issues/127#issuecomment-939180558, or unsubscribe https://github.com/notifications/unsubscribe-auth/AV6X4Y2UBYMUGUU4RW3TILDUF56YDANCNFSM5FRUGQAQ . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.