textrank

Why we included this project

TextRank is the classic graph-based method for pulling the most important sentences and keywords out of a document, and this library is a tidy Python 3 implementation of it. Summarize a text with summarizer.summarize() or extract its key terms with keywords.keywords(); that pair of functions is nearly the whole API. Because it ranks text locally over NumPy and SciPy instead of invoking a language model, it runs fast, gives the same answer every time, and needs no network access, which suits preprocessing steps, indexers, and batch jobs that chew through large volumes of text. A command-line tool handles files directly. It's a focused building block rather than a full summarization platform, but teams that want extractive summaries without model overhead will find it drops into an existing pipeline quickly.