Title: User profile detection based on text mining: technical report no. DCSE/TR-2004-09
Authors: Grolmus, Petr
Issue Date: 2004
Publisher: University of West Bohemia in Pilsen
Document type: zpráva
URI: http://www.kiv.zcu.cz/publications/
Keywords: vyhledávání;uživatelský profil;algoritmus
Keywords in different language: searching;user profile;algorithm
Abstract in different language: The old-fashioned search engine model based usually on Boole´s algebra has become outdated. Current systems tend to meet users (customers) needs more precisely and hence they have to improve their basis. The searching process has to be based on deeper knowledge of individual user preferences - i.e. on user profiles. This work presents an overview of recommender and expert finding systems. It also introduces the ProGen System rising at the University of West Bohemia. For the characteristic phrases selection from document collections is used the STC algorithm. Our system is able to work with multi-lingual document collections. A developed document language recognition is based on stop-word occurrence. We have developed a stemming algorithm combining both the dictionary-based and algorithm-based method. Finally, an outlook for the future work is sketched out.
Rights: © University of West Bohemia in Pilsen
Appears in Collections:Zprávy / Reports (KIV)

Files in This Item:
File Description SizeFormat 
Grolmus.pdfPlný text506,16 kBAdobe PDFView/Open

Please use this identifier to cite or link to this item: http://hdl.handle.net/11025/21604

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.