|Title:||WEB USAGE PATTERN DISCOVERY: THE CASE OF ADDIS ABABA UNIVERSITY OFFICIAL WEB SITE|
|???metadata.dc.contributor.*???:||Dr. Rahel Bekele|
|Abstract:||Web sites are one of the well known communication media where very huge information could be disseminated for visitors‘ access. Since most of the organizations in the world use web sites to communicate their business, smooth running and better service of the web sites is fundamental point for the benefit of users and/or the organization. In order to develop improved web site services, the extent to which users are satisfied on the web site should be computed to evaluate the site. On the first case, the site designers could bring amendments on the site by their expectation of users‘ desire. Secondly, users of the site may be communicated through interview, questionnaire or observation to gather the users‘ trend of the web site usage. Thirdly, researchers could undergo web usage analysis on web server log files of the users‘ requests. In this research, server log of Addis Ababa University web site was utilized to discover web usage pattern using statistical analyzer and data mining applications. Access log file was a target data source as it is expected to show users‘ usage trends. The methodology proposed by Srivastava et al. (2000) was used in the research work which comprised of raw log preparation, pattern discovery and pattern analysis. The well known association mining algorithm, Apriori, was used to extract the common directories that were accessed together. Meanwhile, PredictiveApriori was used in mining common navigational sequences in the web site. Then, the discovered rules were analyzed in consultation with the site map of the web site and the concerned site designers. The research found that there are directories which are frequently visited in the web site. Other findings are also discussed based on which recommendations are forwarded|
|Description:||A Thesis Submitted to the School of Graduate Studies of Addis Ababa University in Partial Fulfillment of the Requirements for the Degree of Master of Science in Information Science|
|Appears in Collections:||Thesis - Information Science|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.