基于改进型相似度的协同过滤算法的研究

Research on Collaborative Filtering Algorithm Based on Improved Similarity

如今的用户面对着大量的信息,从中选取与自己联系密切的相关信息就显得非常困难了。电子商务平台、社区团购平台以及视频平台,面对日益增长的用户数据,需要从中挖掘有利于提高平台效率的信息,因此对用户的个性化推荐提出了更高的要求。各平台都希望能够把握每一个用户的动态信息,实施更加精准的个性化推荐,个性化推荐不仅能够提高平台的效率,也能够为用户带来极致的体验。由于传统的协同过滤算法评价体系中没有考虑到不同用户存在评分的差异,所以在准确性、精确性、差异性等方面仍然需要进一步的提高。该文主要针对在电影评分推荐协同过滤算法中精准化的需求,引入用户差异因子w_(v)^(u),融合原有的皮尔逊相似度计算,从而解决传统的协同过滤算法相似性计算中,针对不同用户具有不同评价体系存在一定偏差的问题。采用Movielens 1m电影评分数据集进行仿真,证明了改进后相似度计算的协同过滤算法能够降低MAE值。

Nowadays,users are faced with a large amount of information,and it is very difficult to select relevant information that has close contact with them.E-commerce platforms,community group buying platforms and video platforms,in the face of ever-increasing user data,need to dig out information that is conducive to improving the efficiency of the platform,so it puts forward higher requirements on personalized recommendations for users.All platforms hope to be able to grasp the dynamic information of each user and implement more accurate personalized recommendations.Personalized recommendations can not only improve the efficiency of the platform,but also bring the ultimate experience to users.Because the traditional collaborative filtering algorithm evaluation system does not take into account the differences in the ratings of different users,the accuracy,precision,and differences still need to be further improved.We mainly aim at the need for precision in the movie rating recommendation collaborative filtering algorithm,introducing user difference factors w_(v)^(u),fusing the original person similarity calculations,so as to solve the problem that there is a certain deviation in the traditional collaborative filtering algorithm similarity calculation for different users with different evaluation systems.The simulation on Movielens 1 m movie scoring data set shows that the improved collaborative filtering algorithm for similarity calculation can reduce the MAE value.