An Approach To Find Frequent Pattern From Logs Using Modified Apriori Algorithm

摘要

Web Usage Mining is an application of Data Mining to generates pattern from the logs which is created when user interacts with websites. Web usage mining is about processing or analysing clickstream data. We can analyse these logs and can find out the user interest and can recommend pages based on user interest. In this paper we have applied Modified Apriori Algorithm on web logs to find patterns. Logs are first preprocessed and then modified Apriori is applied to find interesting pattern which can be used to predict the next page visit of user. The Modified Apriori Algorithm is fast as it requires less scan of database than the basic Apriori algorithm. We have used the dummy dataset and find the frequent patterns using Modified Apriori Algorithm. Modified Apriori is faster as compared to basic Apriori Algorithm as it requires less database scan and therefore suitable for many real time applications.