Log Mining Using Generalized Association Rules

• Main Author: Mohd. Helmy, Abd. Wahab
• Format: Thesis
• Language: English; English
• Published: 2004
• Subjects: QA76 Computer software
• Online Access: https://etd.uum.edu.my/1324/1/MOHD._HELMY_B._ABD._WAHAB.pdf; https://etd.uum.edu.my/1324/2/1.MOHD._HELMY_B._ABD._WAHAB.pdf; https://etd.uum.edu.my/1324/

Explosive growth in size and usage of the World Wide Web has made it necessary for Web site administrators to track and analyze the navigation patterns of Web site visitors. To achieve this goal, the use of web mining tool is necessary. Web mining can be defined as the use of data mining techniques to automatically discover and extract information from web documents. Since Data Mining is primarily concerned with the discovery of knowledge and aims to provide answers to questions that people do not know how to ask, it is not an automatic process. Rather one has to exhaustively explores very large volumes of data to determine otherwise hidden relationships. The process extracts high quality information that can be used to draw conclusions based on relationships or patterns within the data. However, data mining technique are not easily applicable to Web data due to problems both related with the technology underlying the Web and the lack of standards in the design and implementation of Web pages. Information collected by the Web servers are kept in the server log is the main source of data for analyzing user navigation patterns. Once logs have been pre-processed and sessions have been obtained, there are several kinds of access pattern mining that can be performed depending on the needs of the analyst. Since the method use in this study relied on relatively simple techniques therefore the information gathered is adequate for real user profile data due to the noise in the data has to be first tackled. In this study, Data Mining techniques known as generalized association rules was used in order to get some insights into website usage pattern. For the purpose of this study, server logs from tutor.com portal were retrieved, pre-processed and analyzed. An important finding from this study is that Mathematics subject generally popular from UPSR, PMR and UPSR levels. On the contrary, arts subjects are not popular to Tutor.com users. The system administrator may consider evaluating the content and the link for such subjects, so that the real problem can be identified.