基于用户关注度的个性化新闻推荐系统

摘要

There were many problems, such as the shift of user concern, data sparse, and system scalability and automatic update capabilities. This paper gave a new personalized recommendation system for news based on user requirements and user-centered. It described user concern using personal preference and situational interest, used Jacobi to measure user similarity and forecasted user concern with similarity-weighted, and then provided ordered news recommendation list for every user. Experimental results show that the recommendation system can effectively resolve above problems, and it is higher accuracy and coverage and better recommendation results.%为满足用户需求,以用户为中心,解决用户关注度不断变化、数据稀疏性、优化时间和空间效率等问题,提出基于用户关注度的个性化新闻推荐系统.推荐系统引入个人兴趣和场景兴趣来描述用户关注度,使用雅克比度量用户相似性,对相似度加权求和预测用户关注度,从而提供给用户经过排序的新闻推荐列表.实验结果表明,推荐系统有效地提高了推荐精准度和覆盖度,改善了系统可扩展性和自动更新能力,具有良好的推荐效果.