Pages

Friday, October 23, 2009

Semantic web is getting somewhere with Open Calais

The OpenCalais Web Service automatically creates rich semantic metadata for the content you write or load up. Using natural language processing (NLP), machine learning and other methods, Calais analyzes documents and finds the entities within it. But, Calais goes well beyond classic entity identification and returns the facts and events hidden within texts as well.

Frankly, I hadn't heard from it and I read that less than two years into the project, 18,000 developers have joined the OpenCalais community and are now processing four million documents per day, according to Tom Tague, OpenCalais Initiative lead, Thomson Reuters. New sites using it are being announced regularly.

How does it work? If I now write about myself, living in Antwerp, working in e-government, driving a secondhand Renault and having taken the train to work in Brussels today, the result would something like this:

No comments: