A machine learning approach towards improving internet search with a question-answering system

This paper introduces a prototype to extract common sense knowledge from the World Wide Web. The prototype combines a search engine with an automated database. It works by extracting information from the enormous amount of documents available on the World Wide Web. Two common examples are that men love women and that women love men (bi-directional relationship) or that boys like toys (unidirectional relationship), whilst toys cannot like boys.

[1]  Rainer Spiegel,et al.  Combining fuzzy rules and a neural network in an adaptive system , 2002, 2002 IEEE World Congress on Computational Intelligence. 2002 IEEE International Conference on Fuzzy Systems. FUZZ-IEEE'02. Proceedings (Cat. No.02CH37291).

[2]  Ralph Grishman,et al.  Hindi-english cross-lingual question-answering system , 2003, TALIP.

[3]  Xiaohua Hu,et al.  From Computational Intelligence to Web Intelligence , 2002, Computer.

[4]  R. Fikes,et al.  Question answering systems : papers from the 1999 AAAI Fall Symposium, November 5-7, North Falmouth, Massachusetts , 1999 .

[5]  Dieter Fensel,et al.  Ontology-Based Knowledge Management , 2002, Computer.

[6]  Jiawei Han,et al.  Data Mining for Web Intelligence , 2002, Computer.

[7]  Rainer Spiegel,et al.  Abstract and associatively based representations in human sequence learning. , 2003, Philosophical transactions of the Royal Society of London. Series B, Biological sciences.

[8]  Susan T. Dumais,et al.  An Analysis of the AskMSR Question-Answering System , 2002, EMNLP.

[9]  Ning Zhong,et al.  In Search of the Wisdom Web , 2002, Computer.

[10]  Sanda M. Harabagiu,et al.  Mining answers from texts and knowledge bases : papers from the 2002 AAAI Symposium, March 25-27, Stanford, California , 2002 .