# blmoistawinde/HarvestText

文本挖掘和预处理工具（文本清洗、新词发现、情感分析、实体识别链接、关键词抽取、知识抽取、句法分析等），无监督或弱监督方法

Repository: https://github.com/blmoistawinde/HarvestText
Canonical: https://ross.abutalabs.com/products/harvesttext
Language: Python
License: MIT
License Family: permissive
Topics: nlp, sentiment-analysis, new-word-discovery, unsupervised, text-summarization, named-entity-recognition, dependency-parser, text-segmentation, text-cleaning, pyhanlp, harvesttext, keyword-extraction, gitee
Last push: 2024-05-13T08:00:37+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2844, "days_push": 842, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2628, forks 340 (observed 2026-08-28T04:07:05.509470+00:00)

## What it is
HarvestText is a Python library for Chinese text mining and preprocessing using unsupervised or weakly supervised methods. It integrates domain knowledge such as entity types and aliases to support tasks like entity linking, sentiment analysis, new word discovery, keyword extraction, dependency parsing, and simple knowledge graph QA.

## Use cases
- find new words and slang in Chinese social media text
- link entity aliases and abbreviations to canonical names
- analyze sentiment of comments and posts
- extract keywords and summarize documents
- build a social network from a novel by tracking character co-occurrence
- extract fact triples from sentences via dependency parsing
- clean URLs, emails, and special symbols from web text
- segment long unsegmented text automatically

## When to choose
- you need lightweight unsupervised Chinese NLP preprocessing without training data
- you want to combine domain dictionaries and entity aliases with segmentation
- you are doing exploratory analysis of Chinese novels, forums, or specialized literature

## When to avoid
- you need state-of-the-art supervised NER or parsing accuracy
- your primary language is English (support is limited)
- you need a production-grade knowledge graph QA system

## Facets
- artifact type: library
- maturity: maintenance
- function: nlp, parser, search-engine, data-science
- domain: text-editors, data-science
- platform: python, cross-platform
- tags: chinese-nlp, entity-linking, new-word-discovery, sentiment-analysis, text-summarization, keyword-extraction, dependency-parsing, unsupervised, text-cleaning, knowledge-extraction, natural-language-processing

## Member repositories
- blmoistawinde/HarvestText (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:05.509470+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:19:42.516544+00:00, confidence not recorded.
  - readme: https://github.com/blmoistawinde/HarvestText (fetched 2026-08-28T04:07:05.509470+00:00, sha 2b199f7dfc75)
  - registry_pypi: https://pypi.org/pypi/harvesttext/json (fetched 2026-08-29T10:02:35.424623+00:00, sha 6f08d891b493)
- Data as of 2026-08-30T08:39:29.467469+00:00.