Research On The Collaborative Filtering Algorithm Of Personalized Recommendation

With the development of Internet technology and the era of big data,the Internet brings convenience to inform people.However,it also becomes increasingly difficult to get what is valuable from the massive information.To address the problem,the personalized recommendation system is one of the effective ways to reduce excessive information.Its principle is to recommend some information in which the users may have potential interests according to their personal preferences or histories recorded on the Internet.Collaborative filtering recommendation algorithm is one of the most popular recommendation technologies.This essay focuses on the collaborative filtering algorithm of personalized recommendation.By referring to the interests of similar users,collaborative filtering estimates their preferences of certain products and makes recommendations.Aiming to increase accuracy and real timing of this algorithm,improved collaborative filtering algorithms are proposed.One is called collaborative filtering with step screening neighbors(SSN-CF).The new system provides three steps to screen similar neighbors.Before each time,a new screening standard will be introduced based on the previous screening.While searching neighbors,SSN-CF uses the Pearson correlation coefficient method at the first step and considers the influence of users' attributes at the second step.So users with great characteristic differences can be filtered.SSN-CF raises the concept of Prefer set at the last step.Users who have graded the target item or similar items are selected in priority to raise the neighbors' fitness towards the recommendations.The other new system is collaborative filtering based on user interest updating(UIU-CF),which considers short-term interest and long-term interest.In the short-term interest module,UIU-CF adds a forgetting function in the similarity calculation to increase the impact factor of short-term neighbors.Meanwhile,considering the long-term interest of the user,UIU-CF establishes a time threshold of common items to mine forlong-term neighbors,and uses the improved Pearson correlation coefficient method of similarity calculation.So that real-time problem can be improved.With experimental imitation,this essay proves that the new algorithms are more accurate than the traditional one.In such way,the recommendations with the use of new algorithms can be more suitable for target users and the performance of the system is improved.