Automatic discovery of similarity relationships through Web mining

Research output: Journal Publications and Reviews (RGC: 21, 22, 62)21_Publication in refereed journalpeer-review

67 Scopus Citations
View graph of relations



Original languageEnglish
Pages (from-to)149-166
Journal / PublicationDecision Support Systems
Issue number1
Publication statusPublished - Apr 2003
Externally publishedYes


This work demonstrates how the World Wide Web can be mined in a fully automated manner for discovering the semantic similarity relationships among the concepts surfaced during an electronic brainstorming session, and thus improving the accuracy of automated clustering meeting messages. Our novel Context Sensitive Similarity Discovery (CSSD) method takes advantage of the meeting context when selecting a subset of Web pages for data mining, and then conducts regular concept co-occurrence analysis within that subset. Our results have implications on reducing information overload in applications of text technologies such as email filtering, document retrieval, text summarization, and knowledge management. © 2002 Elsevier Science B.V. all rights reserved.

Research Area(s)

  • Context sensitive similarity discovery, Data mining, Empirical study, Group decision support systems, Internet, Machine learning, Organizational concept space, Text clustering, Web mining