Using WordNet to Measure the Similarity of Link Texts

Authors:
Andrzej Siemiński
Affiliations:
Institute of Informatics, Wrocław University of Technology, Wroclaw, Poland 50-370
Venue:
ICCCI '09 Proceedings of the 1st International Conference on Computational Collective Intelligence. Semantic Web, Social Networks and Multiagent Systems
Year:
2009

Citing 6
Cited 1

Information retrieval in the World-Wide Web: making client-based searching feasible

Selected papers of the first conference on World-Wide Web
The shark-search algorithm. An application: tailored Web site mapping

WWW7 Proceedings of the seventh international conference on World Wide Web 7
Focused Crawling Using Context Graphs

VLDB '00 Proceedings of the 26th International Conference on Very Large Data Bases
Introduction to Information Retrieval

Introduction to Information Retrieval
Using evolution strategy for cooperative focused crawling on semantic web

Neural Computing and Applications
Cluster-Centric Approach to News Event Extraction

Proceedings of the 2008 conference on New Trends in Multimedia and Network Information Systems

Fast algorithm for assessing semantic similarity of texts

International Journal of Intelligent Information and Database Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

The primary aim of the study is to evaluate the extend to which the introduction of word similarity defined by the WordNet database could improve the link texts similarity assessment. The proper assessment is crucial for focused crawlers. The crawlers need it to select which links are to be followed. The proposed WordNet based semantic similarity algorithm has increased the recall of the link selection process. To mitigate the co-occurring loss of precision an adaptive algorithm for modifying the initial word similarity levels is introduced and evaluated. The proposed algorithms are verified by an experiment.