Toward an ontology-enhanced information filtering agent
ACM SIGMOD Record
Associate A User's Goal: Exhaustivity and Specificity Information Retrieval Using Ontology
Proceedings of the 2006 conference on Advances in Intelligent IT: Active Media Technology 2006
Hi-index | 0.00 |
With ever-increasing number of websites, users facethe challenge of locating, as well as filtering, browsing, andmonitoring information of astronomical magnitude. While thereare many extant search engines that assist users in locatingURLs, they often return an overwhelmingly large number ofinformation sources for a query, and the (time-consuming butperhaps interesting) task of browsing websites rests heavily onusers. This article presents detailed designs of web agents thatassist users in browsing and filtering information in websites. Inengineering web browsing agents (WBAs), the contributions ofthis research include: (1) devising a 3-stage information filteringapproach that determines the relevance of web pages bydetecting evidence phrases (EP) constructed from WORDNET,counting the frequencies of EP and considering the nearnessamong keywords; and (2) devising a relevance metric tomeasure the relatedness of evidence phrases. Favorableexperimental results show that WBAs are successful in filteringrelevant information in many instances. Discussions on howdifferent word senses affect the information filtering approachare also given.