Collaborative Filtering Recommendation Algorithm Based On Attention Degree

The 21 st century is an information age,We are faced with the problem of Information Overload,so it is very difficult to obtain information that is valuable to us.Although the search engine can alleviate this problem to some extent,it is only relative to the user who know what they want.For some users who don't even know what they want,it's easy to get lost in this huge library of information and brush past the information they're interested in.This is the context in which Personalized Recommendation Systems come into being,But as the number of users and information grows,so does the challenge of Personalized Recommendation Systems.This paper introduces the relevant theoretical knowledge of the Personalized Recommendation System,including the background and research status of the Personalized Recommendation System,and the classification and implementation process of the recommendation system.It mainly introduces the Collaborative Filtering recommendation algorithm in the classification of Personalized Recommendation Systems,including the process and categories of Collaborative Filtering algorithm and the challenges faced.After further research on the Collaborative Filtering algorithm,it is found that the algorithm has shortcomings and can be optimized and improved,which is the focus of this paper.Collaborative Filtering Recommendation Algorithm Based On Attention is proposed in this paper,the algorithm is optimized mainly based on the user's user similarity computing of Collaborative Filtering algorithm,adding user interest deviation as weight in user similarity calculation,improve the accuracy of similarity calculation,so as to improve the accuracy of recommendation in the Recommendation System.Finally,This paper verified improved algorithm by experiment,In the experiment,MAE was selected as the experimental standard.The experimental results show that Collaborative Filtering Recommendation Algorithm Based On Attention improves the accuracy of Recommendation System.