New and more timely business statistics based on web scraping

In this project, business statistics are developed based on natural language processing applied to the texts scraped from the websites of Belgian companies.
Business statistics can be produced in this way on a very frequent basis, based on the entire set of companies with a known website.

This project uses a dataset of all companies with a legal entity in Belgium for which a URL is known.

This project uses web scraping to download visible texts from company websites.

This project uses Natural Language Processing and Machine Learning to automatically categorize the visible text.

This project is still ongoing, and there are no shareable results available yet. Once available they will appear on this page.