基于深度神经网络与协同过滤的视频混合推荐算法  

Video Hybrid Recommendation Algorithm Based on Deep Neural Network and Collaborative Filtering

作　　者：王慧[1] 孙德红[1] WANG Hui(School of Information Management,Minnan University of Science and Technology,Shishi fujian 326700,China)

机构地区：[1]闽南理工学院信息管理学院,福建石狮326700

出　　处：《长春工程学院学报（自然科学版）》2022年第3期110-113,共4页Journal of Changchun Institute of Technology：Natural Sciences Edition

基　　金：2019年福建省中青年教师教育科研项目(JAT190860)。

摘　　要：针对传统视频混合推荐算法中视频推荐响应时间较长、混合推荐效果较差的问题,设计基于深度神经网络与协同过滤的视频混合推荐算法。构建视频混合推荐模型,细化视频混合推荐效果,进行视频推荐模型的数据处理,保证模型数据的真实效果;基于深度神经网络与协同过滤计算视频内容的相似性,缩短视频推荐的响应时间;改进视频相似度量指标,进而实现视频混合的精准推荐。通过对比实验验证所设计算法视频混合推荐响应时间更短,推荐效果更佳。Aiming at the problems of long response time and poor effect of traditional video hybrid recommendation algorithm,a video hybrid recommendation algorithm based on deep neural network and collaborative filtering is designed.Build a video hybrid recommendation model,refine the video hybrid recommendation effect,and process the data of the video recommendation model to ensure the real effect of the model data.The similarity of video content is calculated based on deep neural network and collaborative filtering to shorten the response time of video recommendation.Improve the video similarity measurement index,so as to realize the accurate recommendation of video mixing.Through comparative experiments,it is verified that the video hybrid recommendation response time of this method is shorter,the recommendation effect is better,and it is of the great promotion value.

关 键 词：深度神经网络 协同过滤 视频 算法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]