PDF | IBM Research undertook a challenge to build a computer system that could compete at the human champion level in real time on the American TV quiz. Build watson: An overview of DeepQA for the Jeopardy! The DeepQA project ( ) is aimed at illustrating how the advancement and. @article{journals/aim/FerrucciBCFGKLMNPSW10, added-at = {T +}, author = {Ferrucci, David A. and Brown, Eric W. and.

Author: Doukus Dolmaran
Country: Namibia
Language: English (Spanish)
Genre: Business
Published (Last): 15 July 2008
Pages: 352
PDF File Size: 10.83 Mb
ePub File Size: 4.69 Mb
ISBN: 120-1-75766-566-8
Downloads: 35997
Price: Free* [*Free Regsitration Required]
Uploader: Maugal

The operating existing components is not to force them into a hypothesis is that the correct question interpreta- single, common type system, but to have them tion and derived answer s will score higher after map from the LAT to their own internal types.

A Lexical Database for Eng- applications of machine learning, statistical modeling, lish. In the second round, pdoject dollar values it important for players to know what they know are doubled.

After 3 components was essential for our success to date. Light or Photons answers must rhyme with one another. How to Get on Jeopardy! Champion Human Performance at Jeopardy.

Building Watson: An Overview of the DeepQA Project | AI Magazine

Figure 1 shows the relative fre- contest and evaluation. Logical form alignment wxtson from other sources like triple stores. Film of a typical day in the life of the Bea- Answer: In both phases sets of scores shows why.

The Both present very interesting challenges from an most frequent explicit LATs cover less than 50 AI perspective but were put out of scope for this percent of the data.


Leveraging category information is playing chess.

Perfect confidence estimation upper line and no confidence estimation lower line. Determining whether or not a candidate answer Decomposition. The operative goal for primary search to generate candidate answers. William Murdock, Eric Nyberg, John Prager, Nico Schlaefer, and Chris Welty I IBM Research undertook a challenge to build T he goals of IBM Research are to advance computer science a computer system that could compete at the by exploring new ways for computer dedpqa to affect human champion level in real time on the science, business, and society.

Information For Readers For Authors. Figure 3 contains a graph that illustrates account competition for the buzz. When hit by electrons, a phosphor gives off that overlap by typically one word, and the electromagnetic energy in this form.

The DeepQA so on.

Building Watson: An Overview of the DeepQA Project

The team conducted part of this project has highlighted the need to more than independent experiments in 3 take a systems-level approach to research in QA, years — each averaging about CPU hours and and we believe this applies to research in the generating more than 10 GB of error-analysis data. DeepQA is an architecture with an accompany- Integrate shallow and deep knowledge: He has served on W3 working groups, as pro- ti-Strategy, Multi-Question Approach to Question gram cochair of an international semantic web work- Answering.

Rapid experimentation was another critical The architecture and methodology developed as ingredient to our success. Analysis of Strategic Aditya A. Before and After Goes to the Movies Answer: In David Ferrucci is a research staff member and leads thehe earned his Ph. We adopted a machine-learning approach UIMA-AS handles all of the communication, mes- that requires running the system over a set of train- saging, and queue management necessary using ing questions with known answers and training a the open JMS standard.


For the and hypotheses. We found distinct and Answer: Jeopardy prpject are straightforward assertional forms of questions. Supporting evidence may also object of a pardoning. This is because a player may decide to bet big trade-off between precision and percent answered, on Daily Double or Final Jeopardy questions. The process involves four high-level steps: As between 85 percent and 95 percent precision. We addressed this aspect of adapta- formance on blind Jeopardy data.

His background is representation and reasoning. While potentially compelling wwtson a pub- Figure 2 shows a plot of precision versus percent lic contest, a small number of games does not rep- attempted curves for two theoretical systems. Contestants are shown a picture of a B bomber Outer subclue: Virginia and Indiana explicit LATs in the 20, question sample. A further distinc- a system can deliver far higher precision even with tion is that in these historical games the human the same overall accuracy.

Her research focus is the application of machine learning research interests include question answering, semantic techniques to natural language processing tasks. A lytics to evaluate the supporting evidence.

Chess resources and as-is structured knowledge rather Clue: Communications of the ACM 6.

Previous post: