Skip to content

rapid_textrank

License: MIT Python 3.9+ Rust 2021

High-performance TextRank keyword extraction in Rust with Python bindings.

rapid_textrank extracts keywords and key phrases from text up to 10--100x faster than pure-Python implementations, with support for seven algorithm variants and 18 languages.

Key Features

  • Fast -- Rust core with CSR graph format, string interning, and FxHash for 10--100x speedups over pure Python (depending on document size and tokenization).
  • Eight algorithm variants -- BaseTextRank, PositionRank, BiasedTextRank, TopicRank, SingleRank, TopicalPageRank, MultipartiteRank, and SentenceRank.
  • Unicode-aware -- proper handling of CJK and other scripts.
  • 18 languages -- built-in stopword lists from English to Japanese.
  • Dual API -- native Python classes for quick use, plus a JSON interface for pre-tokenized / spaCy input.
  • Configurable -- control damping, window size, POS filtering, phrase length, score aggregation, and more via TextRankConfig.

Install

pip install rapid_textrank

Quick Example

from rapid_textrank import extract_keywords

text = """
Machine learning is a subset of artificial intelligence that enables
systems to learn and improve from experience. Deep learning, a type of
machine learning, uses neural networks with many layers.
"""

keywords = extract_keywords(text, top_n=5, language="en")
for phrase in keywords:
    print(f"{phrase.text}: {phrase.score:.4f}")

machine learning: 0.2341
deep learning: 0.1872
artificial intelligence: 0.1654
neural networks: 0.1432
systems: 0.0891

Documentation

Section What you will find
Getting Started Installation, quick start guide, and grab-and-go recipes
Algorithms How TextRank works and details on each variant
API Reference extract_keywords(), extractor classes, TextRankConfig, JSON interface, spaCy integration
Performance Benchmarks, why Rust is fast, comparison with alternatives
Architecture Pipeline stages, variant composition, debug introspection, production hardening