Giveme5W1H

Why we included this project

For anyone processing news text, the useful output is usually the same handful of facts: who did what, when, where, why, and how. Giveme5W1H extracts phrases answering each of those questions from an article, using Stanford CoreNLP for the language processing underneath. You can run it as a Python 3 library or through a RESTful API, so it fits into pipelines written in any language. It returns ranked candidate answers per question, which helps when you want the best match rather than a single guess. The tool is deliberately narrow: it turns news text into structured event descriptions for things like media monitoring or event detection, but you will need to wire it into your own processing chain rather than expect an end-to-end product.