IBM cognitive toolkit based on Watson

IBM initially developed Watson as an engine that could play the game of Jeopardy. In this game, a human moderator asks questions in a somewhat cryptic manner in natural language. The question is heard by all the participants at the same time. The players can press a buzzer to indicate that they are ready with an answer. The first player to press the buzzer gets the chance to answer the question. Watson was successful in outperforming the Jeopardy world champion in year 2010. As we can see, this process also goes through the Observe | Interpret | Evaluate | Decide cycle. Here is the high-level architecture of IBM Watson as an intelligent machine that can answer questions in natural language:

>
Figure 12.6: ...

Get Artificial Intelligence for Big Data now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.