About LibriText

LibriText (libritext.org) is a small, independent educational site that explains how computers read and analyse text. The articles are aimed at students, analysts and developers who are meeting natural language processing (NLP) for the first time and want the core ideas without a textbook.

The articles

• Introduction to NLP – tokenization, stemming and lemmatization, part-of-speech tagging and named entities.
• Text similarity and semantic search – from word overlap to embeddings and vector search.
• Text classification – turning text into features and training a first classifier.
• Language detection – how systems identify a language and the difficulties of multilingual text.
• Keyword extraction and summarization – TF-IDF, RAKE, TextRank, YAKE and extractive vs. abstractive summaries.

What LibriText is not

Earlier versions of this website described an NLP platform, API and installable package. No such product exists, and those claims have been removed. LibriText offers articles only – no API keys, software downloads, consulting or paid plans.

LibriText is not affiliated with LibreTexts or with any NLP library or vendor mentioned in the articles.

How the site is run

The site is maintained independently and supported by advertising (Google AdSense).

Contact

Suggestions for new explainers and corrections are welcome via the contact page. Please also see the privacy policy, terms and disclaimer.