摘要
个性化商品推荐系统是电子商务平台系统的重要组成部分,推荐效率的高低直接影响用户的购物体验和电子商务平台商品交易量的提升。近年来,电子商务平台的交易数据呈海量增长趋势,导致商品推荐的正确率下降、误差增大、效率降低,因此对商品个性化推荐算法的研究分析尤为必要。基于企业级阿里云机器学习PAI平台对商品协同过滤推荐算法进行了jaccard、wbcosine和asymcosine三组不同相似度的测试,结果表明,采用jaccard相似度方式进行协同过滤商品推荐效果更佳。
Personalized product recommendation system is an important part of e-commerce platform system,the recommendation efficiency directly affects the user's shopping experience and the promotion of e-commerce platform commodity transaction volume.In recent years,the transaction data of e-commerce platform shows a trend of massive growth,which leads to the decline of the accuracy rate,the increase of error and the decrease of efficiency.Therefore,it is necessary to study and analyze the personalized product recommendation algorithm.Based on the enterprise level Ali cloud machine learning Pai platform,this paper tests three groups of similarity of collaborative filtering recommendation algorithm:Jaccard,wbcosine and asymcosine.The results show that the effect of collaborative filtering product recommendation using Jaccard similarity method is better.
作者
梁家富
LIANG Jia-fu(Guangzhou Vocational and Technical University of Science and Technology,Guangzhou 510550,China)
出处
《河北软件职业技术学院学报》
2020年第4期22-25,41,共5页
Journal of Hebei Software Institute
关键词
机器学习
协同过滤
商品推荐
machine learning
collaborative filtering
product recommendation