Skip to main content

oseti

circleci coveralls.io pyversion latest version license

Dictionary based Sentiment Analysis for Japanese

INSTALLATION

$ pip install oseti

USAGE

import oseti

analyzer = oseti.Analyzer()
analyzer.analyze('天国で待ってる。')
# => [1.0]
analyzer.analyze('遅刻したけど楽しかったし嬉しかった。すごく充実した!')
# => [0.3333333333333333, 1.0]

analyzer.count_polarity('遅刻したけど楽しかったし嬉しかった。すごく充実した!')
# => [{'positive': 2, 'negative': 1}, {'positive': 1, 'negative': 0}])
analyzer.count_polarity('そこにはいつもと変わらない日常があった。')
# => [{'positive': 0, 'negative': 0}]

analyzer.analyze_detail('お金も希望もない!')
# => [{'positive': [], 'negative': ['お金-NEGATION', '希望-NEGATION'], 'score': -1.0}])
analyzer.analyze_detail('お金がないわけではない')
# => [{'positive': ['お金'], 'negative': [], 'score': 1.0}]

# Applying user's dictionary
analyzer = oseti.Analyzer(word_dict={'カワイイ': 'p', 'ブサイク': 'n'},
                          wago_dict={'イカ する': 'ポジ', 'まがまがしい': 'ネガ'})
analyzer.analyze_detail("カワイイ")
# => [{'positive': ['カワイイ'], 'negative': [], 'score': 1.0}]
analyzer.analyze_detail("ブサイクだ")
# => [{'positive': [], 'negative': ['ブサイク'], 'score': -1.0}]
analyzer.analyze_detail("まがまがしい")
# => [{'positive': [], 'negative': ['まがまがしい'], 'score': -1.0}]
analyzer.analyze_detail("イカすよ")
# => [{'positive': ['イカ する'], 'negative': [], 'score': 1.0}]

ACKNOWLEDGEMENT

This module uses 日本語評価極性辞書(用言編)ver.1.0 and 日本語評価極性辞書(名詞編)ver.1.0

  • 小林のぞみ,乾健太郎,松本裕治,立石健二,福島俊一. 意見抽出のための評価表現の収集. 自然言語処理,Vol.12, No.3, pp.203-222, 2005. / Nozomi Kobayashi, Kentaro Inui, Yuji Matsumoto, Kenji Tateishi. Collecting Evaluative Expressions for Opinion Extraction, Journal of Natural Language Processing 12(3), 203-222, 2005.

  • 東山昌彦, 乾健太郎, 松本裕治, 述語の選択選好性に着目した名詞評価極性の獲得, 言語処理学会第14回年次大会論文集, pp.584-587, 2008. / Masahiko Higashiyama, Kentaro Inui, Yuji Matsumoto. Learning Sentiment of Nouns from Selectional Preferences of Verbs and Adjectives, Proceedings of the 14th Annual Meeting of the Association for Natural Language Processing, pp.584-587, 2008.

CHANGES

0.4.1 (2022-08-21)

  • Enable user’s dictionary

  • Solve install issue

  • PEP8

0.3.1 (2022-08-17)

  • Improve dictionaries

  • Use Bunkai module instead of sengiri

0.2 (2019-10-07)

  • Add new methods count_polarity (thanks @yskn67) and analyze_detail

  • Drop support Python 3.3

  • Use sengiri package

  • Support parallel negation (e.g. お金も希望もない is negative)

  • Support “arujanai” expression (e.g. お金があるじゃない is positive)

0.1.2 (2019-02-18)

  • Fix missing CHANGES.rst

0.1.1 (2019-02-13)

  • Set install_requires

0.1 (2019-02-12)

  • First release

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

oseti-0.4.1.tar.gz (74.5 kB view details)

Uploaded Source

File details

Details for the file oseti-0.4.1.tar.gz.

File metadata

  • Download URL: oseti-0.4.1.tar.gz
  • Upload date:
  • Size: 74.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.1 CPython/3.10.6

File hashes

Hashes for oseti-0.4.1.tar.gz
Algorithm Hash digest
SHA256 59896e7868fde1e2af335762161ea871ec6f289d690045898743a53256d8d9ce
MD5 c45765e3e8a9f8a305129a0147262464
BLAKE2b-256 83649bbe2a6e84c7e544373a38443c6b2a2506f9f37af6a2232761ad924ef7a5

See more details on using hashes here.

Release history Release notifications | RSS feed

0.4.3.1

2 files

0.4.3

2 files

0.4.2

1 file

This release

0.4.1 This release

1 file

0.4

1 file

0.3.1

1 file

0.3

1 file

0.2

1 file

0.1.2

1 file

0.1.1

1 file

0.1

1 file

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page