Implementing advanced cleaning and end-user interpretability technologies in Web log mining

Z. Pabarskaite1
1School of Computing, Information Systems and Mathematics, South Bank University, London, UK

Tóm tắt

Two new approaches to Web log mining are presented. Novel advanced cleaning improves Web log mining results. Improved filtering removes pages with no links from other pages. In the data visualisation phase, technical representations of Web pages are replaced by user attractive text interpretations. Experiments with the real world problems showed that the proposed techniques significantly increase the quality and usefulness of Web log mining results.

Từ khóa

#Cleaning #Data visualization #Data mining #Web mining #Web page design #Web server #Data analysis #Information systems #Mathematics #Filtering

Tài liệu tham khảo

10.1109/TAI.1997.632303 banissi, 2000, Information visualization, Encyclopedia of Computer Science and Technology, 42 balabanovic, 1995, An adaptive agent for automated web browsing, Visual Communication and Image Representation, 6 fayyad, 1996, Kdd for science data analysis: Issues and examples, Advances in Knowledge Discovery and Data Mining faulstich, 1998, Wum: A tool for web utilization analysis, EDBT Workshop WebDB98 faulstich, 1999, A data mining analyzing the navigational behaviour of web users, Workshop on Machine Learning User Modelling of the ACAI'99 International Conf 10.1109/DEXA.1998.707429 pitkow, 1994, Webviz: A tool for world-wide web access log analysis, First International World Wide Web Conference jain, 1997, Web mining Pattern discovery from world wide web transactions