Personalization recommendation service in enterprise information portal

Collaborative filtering algorithm is one of the most successful technologies for building recommender systems, and is extensively used in personalized portal. However, existing collaborative filtering algorithms do not consider the change of user interests. For this reason, the systems may recommend unsatisfactory items when user´s interest has changed. To solve this problem, by anglicizing and collecting user´s information and behavior, proposed and established “user-page” matrix as a collaborative filtering algorithm interest matrix, while using the improved cosine similarity collaborative filtering algorithm to calculate the similarity of user interest, and take the initiative to recommend relevant content to users, and the improved algorithm has obviously improved on recommendation accuracy.