Building Watson: An Overview of the DeepQA Project
Создание Watson: обзор проекта DeepQA
2010-09-01
SCID: 54.1/zghbbzsj
Discuss with AI
DeepQA architectureIBM WatsonJeopardy question answeringautomatic contestantreal-time question answering
Figures from the paper
Abstract (AI)
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 show, Jeopardy. The extent of the challenge includes fielding a real‐time automatic contestant on the show, not merely a laboratory exercise. The Jeopardy Challenge helped us address requirements that led to the design of the DeepQA architecture and the implementation of Watson. After three years of intense research and development by a core team of about 20 researchers, Watson is performing at human expert levels in terms of precision, confidence, and speed at the Jeopardy quiz show. Our results strongly suggest that DeepQA is an effective and extensible architecture that can be used as a foundation for combining, deploying, evaluating, and advancing a wide range of algorithmic techniques to rapidly advance the field of question answering (QA).
Key Findings
1
After three years of research and development by approximately 20 core researchers, Watson reached human expert levels in precision, confidence, and speed.
2
IBM developed Watson as a real-time automatic contestant capable of competing at human-champion level on the Jeopardy quiz show.
3
The Jeopardy Challenge drove requirements that shaped the DeepQA architecture and Watson’s implementation, extending beyond a laboratory demonstration.
4
The results indicate that DeepQA provides an effective, extensible foundation for combining, deploying, evaluating, and advancing diverse question-answering algorithms.
Research Object
Watson (the DeepQA question-answering computer system designed to compete on Jeopardy)
Research Subject
Watson's human-expert-level question-answering performance, including precision, confidence, speed, and the extensibility of the DeepQA architecture
Publication Details
Publication Date
2010-09-01
Journal
Publisher
ISSN
Cited by
1553
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai2
Cited by9
Deep Reinforcement Learning: A Brief Survey2017
DBpedia – A large-scale, multilingual knowledge base extracted from Wikipedia2015
A Review of Relational Machine Learning for Knowledge Graphs2015
Deep learning in bioinformatics2016
AI-based chatbots in customer service and their effects on user compliance2020
On-chip photonic synapse2017
RocketQA: An Optimized Training Approach to Dense Passage Retrieval for Open-Domain Question Answering2021
A Survey on Application of Knowledge Graph2020
Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations2017