融合电影属性和交互信息的电影推荐算法研究  

Research on Movie Recommendation Algorithm Combining Movie Attributes and Interactive Information

作　　者：任建宝 邹锟 周元元[2] REN Jianbao;ZOU Kun;ZHOU Yuanyuan(School of Electronic and Information Engineering,Anhui Jianzhu University,Hefei 230601,China;School of Electronic Information and Integrated Circuits,Hefei Normal University,Hefei 230601,China)

机构地区：[1]安徽建筑大学电子与信息工程学院,安徽合肥230601 [2]合肥师范学院电子信息与集成电路学院,安徽合肥230601

出　　处：《苏州市职业大学学报》2024年第1期66-70,共5页Journal of Suzhou Vocational University

基　　金：安徽省高校科研重点项目(2022AH052131)。

摘　　要：作为一种信息过滤系统,电影推荐系统通过分析目标用户的历史数据,向用户推荐他们可能会喜欢的电影。为了提高电影推荐系统的推荐质量,提出一种融合电影属性和交互信息的电影推荐算法U-LFM。该算法首先针对用户评分矩阵的稀疏性问题,使用传统的LFM算法进行矩阵填充;其次针对推荐影响因素的单一性问题,通过融合电影属性以及用户交互信息来增加电影推荐系统的多样性;最后通过实验对比其他几种电影推荐算法在同等条件下的查准率和召回率,验证了U-LFM具有较高的推荐质量。As an information filtering system,the movie recommendation system analyzes the historical data of target users to recommend movies that they may like.In order to improve the recommendation quality of movie recommendation systems,this paper proposes a movie recommendation algorithm U-LFM which combines movie attributes and interactive information.The algorithm first tackles the sparsity problem of the user rating matrix by using the traditional LFM algorithm to fill the matrix.Secondly,to address the issue of singularity in the influencing factors of recommendation,it combines movie attributes and user interactive information to increase the diversity of movie recommendation system.Finally,by comparing the precision and recall rate of other movie recommendation algorithms under the same registration conditions through experiments,it is verified that U-LFM has a high quality of movie recommendation.

关 键 词：推荐算法 协同过滤 隐语义模型 矩阵 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]