Adaptable Web Log Mining from Web Server Logs using Data Preprocessing

With the abundant use of Internet and constant growth of users, the World Wide Web has a huge storage of data and these data serves as an important medium for the getting information of the users access to web sites which are data stored in Web server Logs. Today people are interested in analyzing logs file as they show actual usage of web site. But the data is not accurate so preprocessing of Web log files are essential then after that data are suitable for knowledge discovery or mining tasks. Web Usage or Log Mining, a part of Web mining and application of data mining is used for automatic discovery of patterns in clickstreams and associated data collected or generated as a result of user interactions with one or more Web Sites. We present a comparison study of using enhanced version of decision tree algorithm C4.5 and Naive Bayesian Classification algorithm for identifying interested users. Experimental results conducted shows that the performance metric i.e., time taken and memory to classify the web log files are more efficient when compared to existing C4.5 algorithm.