Skip to main content

概要

変数置き換えモデルを用いた英日両文に適用可能なリーダビリティ判定ツールです。
字種分割にはdivide-char-typeを, 音節数計算にはcount-syllableを使用しています。
戻り値は全体、段落ごと、センテンスごとのリーダビリティ値が取得できるようにしています。

変数置き換えモデルの指標

jFRE = 206.835-(1.015×ASL)-(84.6×ASW)
jFKG = (0.39×ASL)+(11.8×ASW)-15.59
jARI = (4.71×ACW)+(0.5×ASL)-21.43
jCLI = (5.88×ACW)-(29.6/ASL)-15.8
jSMOG = 1.031√(30×PS)+3.1291

*ASL = 字種分割語数/センテンス数
*ASW = 音節数・漢字の連なり数/字種分割語数
*ACW = シャノン情報量に基づく重み/字種分割語数
*PS = 英語3音節・漢字3字以上の字種分割語数/センテンス数

シャノン情報量に基づく重みは、英数字(61種類)を1として、ひらがな(88種類)をlog(1/88)/log(1/61)で,カタカナ(141種類)をlog(1/141)/log(1/61)で、漢字(20898種類)をlog(1/20898)/log(1/61)でそれぞれ重み付けする.

評価表

jFREはReading Ease Scoreに照らし合わせて評価します。
jFKG、jARI、jCLI、jSMOGはEstimated Reading Gradeに照らし合わせて評価します。

Reading Ease Score Style Description Estimated Reading Grade Estimated Percent of U.S. Adults (1949)
0 to 30: Very Difficult College graduate 4.5
30 to 50: Difficult 13th to 16th grade 33
50 to 60: Fairly Difficult 10th to 12th grade 54
60 to 70: Standard 8th to 9th grade 83
70 to 80: Fairly Easy 7th grade 88
80 to 90: Easy 6th grade 91
90 to 100: Very Easy 5th grade 93

セットアップ

pip install calculate-readability

アンインストール

pip uninstall calculate-readability divide-char-type count-syllable nltk

使用方法

from calculate_readability import calculate_readability

data = calculate_readability("今日の天気は晴れです。明日は曇りです。\n明後日は雨です。")

print(data["raw_text"])
print(data["text"])
print(data["jfre"])

print(data["break"][0]["text"])
print(data["break"][0]["jfre"])

print(data["break"][0]["sentence"][0]["text"])
print(data["break"][0]["sentence"][0]["jfre"])

論文

別途、論文化、または、学会発表を予定してます。

ライセンス

  • calculate-readability
    • Python Software Foundation License
    • Copyright (C) 2024 Shinya Akagi
  • divide-char-type
    • Python Software Foundation License
    • Copyright (C) 2023-2024 Shinya Akagi
  • count-syllable
    • Python Software Foundation License
    • Copyright (C) 2024 Shinya Akagi
  • nltk
    • Apache License 2.0
    • Copyright (C) 2001-2023 NLTK Project
  • cmudict
    • BSD License
    • Copyright (C) 1998 Carnegie Mellon University

Metadata

Release files for calculate-readability 0.1.2

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for calculate-readability 0.1.2
File Size Uploaded Provenance
calculate_readability-0.1.2.tar.gz 4.6 kB Without provenance Details

Built distribution (wheel)

Table of built distributions (wheels) for calculate-readability 0.1.2
File Interpreter ABI Platform
calculate_readability-0.1.2-py3-none-any.whl Python 3 none any Details

Total release size: 9.4 kB

Release files / calculate_readability-0.1.2.tar.gz

Download URL calculate_readability-0.1.2.tar.gz
Size 4.6 kB
Tags Source
SHA-256 checksum
How to use checksums
d55b5cf74cc5eaae2cf72b3d085bf91f1b72836e3862b18210e8bfb726ada26c
BLAKE2b-256 checksum
How to use checksums
bd33f4017b6f5daa19bb447b29d3f178964533cce3806c2bb7563121be317af4
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/4.0.1 CPython/3.10.15

Release files / calculate_readability-0.1.2-py3-none-any.whl

Download URL calculate_readability-0.1.2-py3-none-any.whl
Size 4.8 kB
Tags Python 3
SHA-256 checksum
How to use checksums
630bdbd23a5e6954a9f7760af2debf1c6e05955fbcad77585b26ea39292d650d
BLAKE2b-256 checksum
How to use checksums
8e752fbf92d7501c817e902a66f380ff5b420e873f2fc2a315726f89a711392a
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/4.0.1 CPython/3.10.15

Security for calculate-readability 0.1.2

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

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