Monday, March 2, 2015

The Linguistics behind IBM Watson

I will be talking about the linguistics behind IBM Watson's Question Answering on March 11 at the DC Natural Language Processing MeetUp. Here's the blurb:

In February 2011, IBM Watson defeated Brad Rutter and Ken Jennings in the Jeopardy! Challenge. Today, Watson is a cognitive system that enables a new partnership between people and computers that enhances and scales human expertise by providing a more natural relationship between the human and the computer. 

One part of Watson’s cognitive computing platform is Question Answering. The main objective of QA is to analyze natural language questions and present concise answers with supporting evidence, rather than a list of possibly relevant documents like internet search engines.

This talk will describe some of the natural language processing components that go into just three of the basic stages of IBM Watson’s Question Answering pipeline:

  • Question Analysis
  • Hypothesis Generation
  • Semantic Types

The NLP components that help make this happen include a full syntactic parse, entity and relationship extraction, semantic tagging, co-reference, automatic frame discovery, and many others. This talk will discuss how sophisticated linguistic resources allow Watson to achieve true question answering functionality.

No comments:

Putting the Linguistics into Kaggle Competitions

In the spirit of Dr. Emily Bender’s NAACL blog post Putting the Linguistics in Computational Linguistics , I want to apply some of her thou...