Introduction to "This is Watson"

From Brede Wiki
Jump to: navigation, search
Paper (help)
Introduction to "This is Watson"
Authors: David A. Ferrucci
Citation: IBM Journal of Research and Development 56 (3.4): missing pages. 2012 May-June
Database(s): Google Scholar cites
DOI: 10.1147/JRD.2012.2184356.
Web: Bing Google Yahoo!Google PDF
Article: BASE Google Scholar PubMed
Restricted: DTU Digital Library
Other: NIF
Format: BibTeX

Introduction to "This is Watson" is about IBM Watson, the open-domain question-answering system.

The paper argues that such systems require advances in areas of

The Watson system (AdaptWatson) consists of "more than 100 core algorithmic components."

Some of the components are:

  • Question analysis to determine lexical answer type
  • Grammatical parser of the question.
  • Collection of content, e.g., from encyclopedias.
  • Building a resource (PRISMATIC) with knowledge from collected content.

[edit] Results

Initial performance was 16% precision@70 (i.e., with 70% questions answered). It rose to 85% Precision@70.

Table 1:

Task State of the art
Entity disambiguation Robust disambiguation of named entities in text
Relation detection A composite kernel to extract relations between entities with both flat and structured features
Textual entailment PKUTM participation TAC 2010 RTE and summarization track

[edit] Related papers

  1. A framework for merging and ranking of answers in DeepQA
  2. Automatic knowledge extraction from documents
  3. Building Watson: an overview of the DeepQA project
  4. Deep parsing in Watson
  5. In the game: the interface between Watson and Jeopardy!
  6. Natural language processing with Prolog in the IBM Watson system
  7. Question analysis: how Watson reads a clue
  8. Relation extraction and scoring in DeepQA
  9. Textual resource acquisition and engineering
  10. Towards the open advancement of question answering systems

[edit] See also

  1. Unstructured Information Management Architecture
Personal tools