Overview¶

The notnews package provides classifiers for soft news based on story text and URL structure for both US and UK news media. We also provide ways to infer the ‘kind’ of news—Arts, Books, Science, Sports, Travel, etc.—for US news media.

Modern Features¶

Traditional ML classifiers - Fast, offline classification using trained models
LLM-based classification - Flexible classification using Claude and OpenAI with custom categories
Web content fetching - Automatically fetch and classify content from URLs

Getting Started¶

Installation - Install notnews with optional LLM dependencies
Quick Start - Quick examples to get you started

API Reference¶

Traditional ML API - Traditional ML-based classifiers (5 functions)
LLM-based API - Modern LLM-based classification with Claude/OpenAI
Command Line Interface - Command line tools for batch processing

Applications & Data¶

Underlying Data¶

For more information about how to get the underlying data for UK model, see here. For information about the data underlying the US model, see here

Applications¶

We use the model to estimate the supply of not news in the US and the UK.

Documentation¶

For more information, please see project documentation.

Authors¶

Suriyan Laohaprapanon and Gaurav Sood

Contributor Code of Conduct¶

The project welcomes contributions from everyone! In fact, it depends on it. To maintain this welcoming atmosphere, and to collaborate in a fun and productive way, we expect contributors to the project to abide by the Contributor Code of Conduct

License¶

The package is released under the MIT License.

Streamlit App: Try notnews online