DerwenAI/pytextrank

Health v2 · maintenance only

Adoption not part of the score

2218 stars · 334 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

PyTextRank is a Python implementation of TextRank and related textgraph algorithms (PositionRank, Biased TextRank, TopicRank) as a spaCy pipeline extension. It provides graph-based ranking for phrase extraction and low-cost extractive summarization of text documents.

Use cases

• extract top-ranked key phrases from a document
• summarize long text documents extractively
• extract keywords from articles with python
• build a spacy pipeline for phrase extraction
• infer concepts from unstructured text into structured representations
• rank phrases in text using graph algorithms

When to choose

• you already use spaCy and want a drop-in pipeline extension
• you need lightweight, dependency-free extractive summarization without training models
• you want graph-based keyword/phrase extraction with multiple algorithm variants

When to avoid

• you need abstractive summarization or generative models
• you don't use spaCy and prefer standalone NLP tooling
• you need state-of-the-art transformer-based keyphrase extraction

Facets

library · maturity active

nlp parser machine-learning machine-learning data-science python textrank spacy-extension phrase-extraction extractive-summarization graph-algorithms keyword-extraction textgraphs natural-language-processing

Member repositories

For agents

markdown · JSON · MCP: product_card(name="DerwenAI/pytextrank")