Masters Theses

Abstract

The World Wide Web has become a very popular marketing medium during the recent years. Many businesses use it extensively to publicize their products. Initially it began with static web sites. However as web sites got bigger and better, their maintenance became problematic. Just as business strategies change, so also will the orgainization of their web sites change. Hence the need of the hour is an adaptive web site (i.e. a web site that can organize itself based upon its usage). For this purpose, web usage mining techniques are used. The characteristics of the user's activity on the web site is obtained from a web server log and this data is preprocessed and later used to modify the web site.

In this thesis, a method for the reorganization of a Web site based on access patterns is proposed. Page access information of the Web users is extracted from Web servers' log files and then organized into sessions which represent episodes of interaction between Web users and the Web server. Using statistical analysis, the information regarding the various pages which comprise the Web site is obtained. This information is then used to categorize pages into two major categories INDEX and CONTENT. After the appropriate categorization is done, the Web site is then processed and rebuilt as per the algorithm proposed. Our experiments on a large real data set show that the method is efficient and practical for Web mining applications. However, for best possible results, prior knowledge of the web site and its potential users should be well understood in order to set appropriate parameters and thresholds--Abstract, p. iiii

Advisor(s)

Yongjiam Fu

Committee Member(s)

Arlan R. Dekock
Raymond Kluezny

Department(s)

Computer Science

Degree Name

M.S. in Computer Science

Publisher

University of Missouri--Rolla

Publication Date

Spring 2001

Pagination

ix, 78 pages

Note about bibliography

Includes bibliographical references (pages 74-77)

Rights

© 2001 Mario Rodney Creado, All rights reserved.

Document Type

Thesis - Open Access

File Type

text

Language

English

Thesis Number

T 7902

Print OCLC #

47769361

Share

 
COinS