Improvement of Personalized Dynamic Service using Independent Component Analysis

Author

Pu, Qiang ; Liu, Qi-He ; Yang, Guo-Wei

Author_Institution

Sch. of Comput. Sci. & Eng., UEST of China, Chengdu

fYear

2006

fDate

13-16 Aug. 2006

Firstpage

2611

Lastpage

2616

Abstract

A system of personalized dynamic service (PDS) in digital library (DL) based on independent component analysis (ICA) was proposed and discussed in detail in order to improve the quality of personalized service to readers. Meanwhile, architectures of general PDS system and the most important subsystem were introduced. The experiments showed that better effects of books´ topic classifying and readers´ interest clustering could be respectively achieved by employing ICA method, and a dynamic mapping relation that improved the quality of PDS, furthermore, could be adaptively generated along with the changes of readers´ personalized interests or books´ topics

Keywords

classification; digital libraries; independent component analysis; text analysis; books; digital library; independent component analysis; personalized dynamic service; reader interest clustering; topic classification; Books; Clustering algorithms; Computer architecture; Computer science; Cybernetics; Independent component analysis; Machine learning; Machine learning algorithms; Samarium; Software libraries; Text categorization; Time factors; Independent component analysis; Personalized dynamic service; Text classification;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Learning and Cybernetics, 2006 International Conference on

Conference_Location

Dalian, China

Print_ISBN

1-4244-0061-9

Type

conf

DOI

10.1109/ICMLC.2006.258913

Filename

4028504