Digital Library

cab1

 
Title:      MAIN RETRIEVAL SYSTEMS’ PARAMETERS ANALYZE
Author(s):      João Ferreira , Alberto Rodrigues Da Silva , José Delgado
ISBN:      972-99353-6-X
Editors:      Nuno Guimarães and Pedro Isaías
Year:      2005
Edition:      1
Keywords:      Vectorial, Links Analyses, Retrieval Information, Classification, and Probabilistic.
Type:      Full Paper
First Page:      344
Last Page:      351
Language:      English
Cover:      cover          
Full Contents:      click to dowload Download
Paper Abstract:      We explore retrieval effectiveness of vectorial, link analyses, probabilistic and classification retrieval system parameters and compare the different performance in a controlled environment (using WT10g collection from TREC). Main retrieval parameters are captured from related to: (1) retrieval methods (e.g., vectorial, link analyses, probabilistic and classification); (2) main internal system parameters (e.g., query length, URL length, phrase, feedback, index); (3) and also internal system parameters combination (e.g., combination of different query and URL lengths, phrase, feedback and index) and we concluded about most important parameters, retrieval method performance and that combination can improve results. We analyses many cases from around 500 retrieval systems using our own modular platform, WebSearchTester.
   

Social Media Links

Search

Login