Use of Text Mining Methods in a Digital Library

Use of Text Mining Methods in a Digital Library

The article deals with use of Itemsets classifier based on inductive machine learning in the context of digital library environment. We provide a brief description of a real-world digital library implemented at a power utility. Its implementation and operating experience have motivated our research in inductive machine learning methods for text mining described in the paper. Being inspired by mining of association rules, we have developed a new categorization method named “Itemsets classifier”. By performing various experiments we have proved its ability to surpass some well-known categorization methods, both in terms of precision/recall and efficiency. As the task of classification is closely related to clustering, we have integrated the principles of Itemsets method into a new document-clustering algorithm as well. We are also presenting other Itemsets classifier applications in unsolicited mail filtering and enhancement of the Naïve Bayes classifier. Main ideas and experimental results are presented in the paper.Copyright for the full paper: Verlag für Wissenschaft und Forschung, VWF, Berlin, Germany.

Keywords: classification, clustering, categorization, classifier, spam filter, machine learning

Year: 2002

Download: download Full text [56 kB]

Authors of this publication:


Jiří Hynek


Phone: +420 603492837
E-mail: jhynek@kiv.zcu.cz
WWW: http://www.kiv.zcu.cz/staff/osobni.php?id_osoby=147&lang=EN

Jiri, a co-founder of the Text-Mining Research Group, works as a lecturer at the Dept. of Computer Science and Engineering. His research interests include machine learning and language-related problems. Jiri’s teaching activity is focused on good writing style and technical writing in general.

Karel Ježek


Phone:  +420 377632475, 377632400
E-mail: jezek_ka@kiv.zcu.cz
WWW: http://www-kiv.zcu.cz/~jezek_ka/

Karel is a group coordinator and a supervisor of PhD students working at research projects of this Group.

Related Projects:


Project

Document Classification

Authors:  Jiří Hynek, Karel Ježek, Michal Toman, Roman Tesař, Zdeněk Češka, Petr Grolmus
Desc.:Use of inductive machine learning methods in classification of short text documents.