Effective web log mining and online navigational pattern prediction

The web has become the world's largest repository of knowledge. Web usage mining is the process of discovering knowledge from the interactions generated by the user in the form of access logs, cookies, and user sessions data. Web Mining consists of three different categories, namely Web Content Mining, Web Structure Mining, and Web Usage Mining (is the process of discovering knowledge from the interaction generated by the users in the form of access logs, browser logs, proxy-server logs, user session data, cookies). Accurate web log mining results and efficient online navigational pattern prediction are undeniably crucial for tuning up websites and consequently helping in visitors’ retention. Like any other data mining task, web log mining starts with data cleaning and preparation and it ends up discovering some hidden knowledge which cannot be extracted using conventional methods. After applying web mining on web sessions we will get navigation patterns which are important for web users such that appropriate actions can be adopted. Due to huge data in web, discovery of patterns and there analysis for further improvement in website becomes a real time necessity. The main focus of this paper is using of hybrid prediction engine to classify users on the basis of discovered patterns from web logs. Our proposed framework is to overcome the problem arise due to using of any single algorithm, we will give results based on comparison of two different algorithms like Longest Common Sequence (LCS) algorithm and Frequent Pattern (Growth) algorithm. Keywords— Web Usage Mining, Navigation Pattern, Frequent Pattern (Growth) Algorithm. ________________________________________________________________________________________________________