A devised framework for content recommendation system using collaborative log mining

Internet is proving to play such an important role in our everyday life that it is almost impossible to survive without it. The World Wide Web (WWW) has inclined a lot to both users (visitors) as well as the web site owners. Enormous growth of World Wide Web increases the complexity for users to browse effectively and efficiently. Users visit a web site with a quench of getting useful information he/she is interested in. So as to satisfy user´s objective and goal of searching web sites, betterment in web site design, web server activities are required to be changed as per users´ interests. To achieve this analysis of user access pattern, which are captured in the form of log files is required, known as Web Usage Mining (WUM). Content recommendation system suggests and assists in selecting the content from wide and complex search space, which match with visitors interest, and which are unknown to them. Amount and type of interaction with web page captured at client side is an indicative measure of appropriateness of the content presented. Here in this paper, we provide detailed survey on various approaches for content recommendation and work done so far in this area, and proposed an hybrid approach that considers data gathered at client side along with web server´s web log data, which will be used collaboratively to recommend a content.