search

luisgustavoneves / happy

MIT License
73 stars 7 forks source link

Unhappiness #3

Open luisgustavoneves opened 10 years ago

luisgustavoneves commented 10 years ago

It is easy to have the program count only happy words. For instance h.hi(text, 0, 7) will only consider words with scores over 7 and will work much more like conventional sentiment analysis software. To account for unhappiness is more complicated. h.hi(text, 3, 10) can be used, but words with scores near 3 will end up having more weight than words with lower scores, what doesn't seem reasonable. I wonder what can be a good way to adjust scores for unhappiness. Perhaps subtracting 3 from words with scores lower than that and counting the negatives.

Trueflash commented 8 years ago

@luisgustavoneves sir , Can be used for Chinese?

luisgustavoneves commented 8 years ago

Yes. But you have to download the target data, which is also available in Chinese. Setup

First download files for target language:

http://www.uvm.edu/storylab/share/papers/dodds2014a/data.html

Then configure your language at the beginning of the code:

LANG = "target_language"

2016-07-27 22:41 GMT-03:00 Trueflash notifications@github.com:

@luisgustavoneves https://github.com/luisgustavoneves sir , Can be used for Chinese?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/luisgustavoneves/happy/issues/3#issuecomment-235773099, or mute the thread https://github.com/notifications/unsubscribe-auth/AC0sShvJA4jWLzCTkxvRfTms7Mf1ySGFks5qaAjGgaJpZM4CS74O .