DERIVING USER ACCESS PATTERNS AND MINING WEB COMMUNITY WITH WEB-LOG DATA FOR PREDICTING USER SESSIONS WITH PAJEK

• Main Authors: S. Balaji, S. Sasikala
• Format: Article
• Language: English
• Published: ICT Academy of Tamil Nadu 2012-10-01
• Series: ICTACT Journal on Soft Computing
• Subjects: Web Log; Mining; Clustering; Social Network Analysis
• Online Access: http://ictactjournals.in/paper/IJSC_Vol3_Iss1_Paper3_415_419.pdf

Web logs are a young and dynamic media type. Due to the intrinsic relationship among Web objects and the deficiency of a uniform schema of web documents, Web community mining has become significant area for Web data management and analysis. The research of Web communities extents a number of research domains. In this paper an ontological model has been present with some recent studies on this topic, which cover finding relevant Web pages based on linkage information, discovering user access patterns through analyzing Web log files from Web data. A simulation has been created with the academic website crawled data. The simulation is done in JAVA and ORACLE environment. Results show that prediction of user session could give us plenty of vital information for the Business Intelligence. Search Engine Optimization could also use these potential results which are discussed in the paper in detail.