Personalized content optimization using tensor segmentation

World Wide Web (WWW) is a huge collection of several web sites and links and the volume of data stored on the internet increases day by day. So the size and complexity of many web sites grow along with it. Finding relevant information on large website can be difficult and time consuming process. A website can be personalized to help users to find the exact content what they need from huge source of data. The goal of personalization is to deliver the content what the user need, without asking them explicitly. Existing personalization techniques such as Collaborative filtering uses recommender systems where users' additional effort is involved. In this paper Tensor Segmentation is used to engage user in the manifestation of user details and product features. Behavior Based Supervised Clustering algorithm is proposed to segment users based on their relevant features. It analyses the click behavior from web search and provides the most accurate optimization of web data without involving significant effort of user.