Emotion analyzer for Japanese
Project description
pymlask
pymlask is a Python version of ML-Ask (eMotive eLement and Expression Analysis system)
For details about ML-Ask, See http://arakilab.media.eng.hokudai.ac.jp/~ptaszynski/repository/mlask.htm
See also http://qiita.com/yukinoi/items/ef6fb48b5e3694e9659c (in Japanese)
Contributions are welcome!
Installation
Modified dictionary version
pip install pymlask
ML-Ask Original dictionary version (same as Ptaszynski’s Perl version)
pip install git+https://github.com/ikegami-yukino/pymlask@original
Example
from mlask import MLAsk
emotion_analyzer = MLAsk()
emotion_analyzer.analyze('彼のことは嫌いではない！(;´Д`)')
# => {'text': '彼のことは嫌いではない！(;´Д`)',
# 'emotion': defaultdict(<class 'list'>,{'yorokobi': ['嫌い*CVS'], 'suki': ['嫌い*CVS']}),
# 'orientation': 'POSITIVE',
# 'activation': 'NEUTRAL',
# 'emoticon': ['(;´Д`)'],
# 'intension': 2,
# 'intensifier': {'exclamation': ['！'], 'emotikony': ['´Д`', 'Д`', '´Д', '(;´Д`)']},
# 'representative': ('yorokobi', ['嫌い*CVS'])
# }LICENSE
The BSD 3-Clause License
Copyright
ML-Ask Python: The BSD 3-Clause License (c) 2017 Yukino Ikegami. All Rights Reserved.
ML-Ask (original): The BSD 3-Clause License (c) 2007-2013, Michal Ptaszynski, Pawel Dybala, Rafal Rzepka, Kenji Arakii. All Rights Reserved.
CHANGES
0.3.2 (2019-07-09)
Fix bugs about emotion pattern matching (thanks @brunotoshio)
Fix bug about activation count (thanks @brunotoshio)
0.3.1 (2019-05-22)
Use MeCab.Tagger().parse() instead of MeCab.Tagger().parseToNode
0.3 (2019-05-17)
The 712 emotional words from Kanjou kotoba erabi jiten (感情ことば選び辞典) are newly added
Support Python 3.7
Unsupport Python 2.6 and 3.3
0.2.5 (2017-09-14)
Fix bugs about MeCab (thanks @Kensuke-Mitsuzawa)
Delete install_requires
0.2.4 (2017-03-01)
Fix many bugs
Add some emotional words
Delete invalid words
Correct typo
0.2.1 (2017-02-23)
Add 67 emotional words
0.2 (2017-02-22)
Support Python 2.X
Add 52 emotional words
Fix bug
0.1.1 (2017-02-15)
Delete debug print (thanks @ssirai)
0.1 (2017-02-10)
First release.
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
