一种新型的Flink电子商务实时推荐系统  

A new Flink E-Commerce Real-Time Recommendation System

作　　者：马腾 牛少彰[1,2] 史成洁 MA Teng;NIU Shao-zhang;SHI Cheng-jie(School of Computer Science,Beijing University of Posts and Telecommunications,Beijing 100876,China;Institute of Information Engineering,Chinese Academy of Sciences,Beijing 100093,China)

机构地区：[1]北京邮电大学计算机学院,北京100876 [2]中科院信息工程研究所,北京100093

出　　处：《新一代信息技术》2021年第4期22-28,共7页New Generation of Information Technology

基　　金：国家自然科学基金(项目编号:61370195,61070207)。

摘　　要：推荐算法是数据挖掘领域中应用最广泛的技术之一,目前的推荐算法主要针对静态数据,缺乏对流式数据的适应性,和实时变化性的要求。本文提出了一种实时推荐算法,该算法每隔一定时间间隔离线更新商品特征矩阵和相似度矩阵,可以在用户实时评分之后,立刻根据用户最近一段时间的评分,产生实时推荐列表。并且在新的分布式流计算框架Flink上实现了一个电商推荐系统,在renttherunway数据集上进行了实验。实验结果表明,该算法能够在保证推荐准确率和可扩展性的同时,满足实时性和变化率的要求。The recommendation algorithm is one of the most widely used technologies in the field of data mining.The current recommendation algorithm is mainly aimed at static data,lacking adaptability to streaming data and real-time variability requirements.This paper proposes a real-time recommendation algorithm.The algorithm updates the product feature matrix and the similarity matrix offline at regular intervals.It can generate a real-time recommendation list based on the user's recent rating after the user's real-time rating.And implemented an e-commerce recommendation system on the new distributed stream computing framework Flink,and conducted experiments on the renttherunway dataset.Experimental results show that the algorithm can meet the requirements of real-time and change rate while guaranteeing the accuracy and scalability of recommendation.

关 键 词：计算机科学与技术 推荐系统 实时 电商 Flink 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]