#6666th of 69 in Data and scraping for AIFor data and documents, data analysis
scattertext
Python library that plots distinguishing terms between text corpora as interactive HTML
- Stars
- 2.4k
- License
- Apache-2.0
- Last commit
- Jul 2026
- Last release
- Mar 2017
- Language
- Python
2,322 on 25 Apr 20262,351 on 14 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Scattertext is a Python library that scores which words and phrases are characteristic of one document category versus another, then writes a standalone interactive HTML scatter plot with non-overlapping point labels. It includes scoring methods such as Scaled F-Score, Cohen's d, Cliff's Delta and Bi-Normal Separation, plus dispersion metrics. A `scattertext` command-line tool handles CSV input without writing Python.
Who it is for: Data scientists and linguists comparing vocabulary across text categories
Strengths
- Outputs a self-contained HTML file with zoom, click-through term stats and non-overlapping labels
- Many term scoring options: Scaled F-Score, Cohen's d, Cliff's Delta, BNS, dispersion measures
- Runs without spaCy using a regex-based whitespace tokenizer
- CLI analyzes a CSV with text and category columns, no Python needed
Weaknesses
- Classical corpus statistics; no LLM integration, so not an LLM-era tool
- Last tagged release is 2017-03-13 although commits continue
- Full feature set needs optional packages: spacy, gensim, empath, umap-learn, jieba, flashtext
- README recommends Chrome or Safari for viewing output; docs are a work in progress
What it needs
- no GPU
- Needs Python 3.11+, spacy (optional), gensim (optional), empath (optional), umap-learn (optional), jieba (optional), flashtext (optional), astropy (optional)
What changed
From the release notes. No flag means the note does not say so, not that the upgrade is safe.
Also in Data and scraping for AI
See all 69| Rank | Project | Score |
|---|---|---|
| 1 | mineruParses PDFs, images and Office files into Markdown and structured JSON | 91 out of 100 |
| 2 | doclingPython library that converts PDFs, Office files and more into structured text | 91 out of 100 |
| 3 | opendataloader-pdfPDF parser that outputs Markdown, JSON with bounding boxes, and Tagged PDF | 91 out of 100 |
| #4 | agent-reachCLI that gives coding agents read access to login-gated and Chinese web platforms | 88 out of 100 |
| #5 | markitdownPython library and CLI that converts office files, PDFs and audio to Markdown | 87 out of 100 |