Build Your Own Watson with Open Source Software

The key analytical technology underlying IBM’s Watson, winner of the Jeopardy! contest against two previously unbeaten contestants, is an open source program called UIMA, for Unstructured Information Management application.

Speaking at the 2011 High Performance Computing for Wall Street, Edward Epstein of IBM Research, explained that Watson was designed to work with natural language content and deep analysis drawing on all the available content. The big idea behind Watson is evidence-based reasoning over natural language content. UIMA, which IBM developed and turned over to Apache Software Foundation, is composed of hundreds of complex algorithms. UIMA helps solve the interoperability issues to deliver just the data needed.

As the company explains on its web site:

“Unstructured Information Management Architecture (UIMA) is an open framework for building analytic applications – to find latent meaning, relationships and relevant facts hidden in unstructured text.”