search

ikegami-yukino / pymlask

Emotion analyzer for Japanese text
Other
114 stars 26 forks source link

Activation issue #11

Closed brunotoshio closed 5 years ago

brunotoshio commented 5 years ago

Hi,

I think I found a bug related to the activation. 

from mlask import MLAsk

ma = MLAsk()
ma.analyze('嫌です')    # => activation = ACTIVE, but it should be NEUTRAL

The problem is that you are counting the number of letters A after the regular expression substitutions without taking into account that the word NEUTRAL also has a letter A. Therefore, the emotion iya, which belongs to NEUTRAL, is being counted as an A (active) too.

ikegami-yukino commented 5 years ago

Thank you for reporting this issue. I resolved it 😃