Learning while filtering documents
- 1 August 1998
- proceedings article
- Published by Association for Computing Machinery (ACM)
- p. 224-231
- https://doi.org/10.1145/290941.290998
Abstract
This paper examines the problems of learningqueries and dissemination thresholds from relevancefeedback in a dynamic information filtering environment.It revisits the EG algorithm for learning queries, identifyingseveral problems in using it reliably for informationfiltering, and providing solutions. It also presents a newalgorithm for learning dissemination thresholds automatically,from the same relevance feedback information usedto learn queries.1 IntroductionAn information...Keywords
This publication has 8 references indexed in Scilit:
- Training algorithms for linear text classifiersPublished by Association for Computing Machinery (ACM) ,1996
- Incremental relevance feedback for information filteringPublished by Association for Computing Machinery (ACM) ,1996
- Document filtering with inference networksPublished by Association for Computing Machinery (ACM) ,1996
- The design of a high performance information filtering systemPublished by Association for Computing Machinery (ACM) ,1996
- Optimization of relevance feedback weightsPublished by Association for Computing Machinery (ACM) ,1995
- Personalized information deliveryCommunications of the ACM, 1992
- Retrieving highly dynamic widely distributed informationPublished by Association for Computing Machinery (ACM) ,1989
- The importance of SDI for current awareness in fields with severe scatter of informationJournal of the American Society for Information Science, 1979