User Interest Change-adaptive Recommendation model based on Social Tagging

Many recommendation algorithms based on social tagging ignore the change and repeatability of user interests. In order to solve these problems, a new user interest adaptive recommendation model is proposed, which efficiently combines exponential forgetting-based data weight and time windows. The new model not only highlights the importance of recent interest, but also stresses the recurring early data. The nearest neighbor set can be gained according to exponential offset tag vectors, and then make recommendations by calculating similarity between resource set of the nearest neighbors and that of the target user which is tipped within time windows. The simulation experiments show that the proposed algorithm for recommendation has high quality of precision to some extent.