《计算机应用》唯一官方网站 ›› 2022, Vol. 42 ›› Issue (11): 3506-3512.DOI: 10.11772/j.issn.1001-9081.2021111992

• CCF 2021中国数字服务大会 • 上一篇    

基于云‒端融合的个性化推荐服务系统

韩佳良1, 韩宇栋1, 刘譞哲1, 赵耀帅2,3, 冯迪2,3()   

  1. 1.高可信软件技术教育部重点实验室(北京大学), 北京 100871
    2.中国民航信息网络股份有限公司, 北京 101318
    3.中国民用航空局 民航旅客服务智能化应用技术重点实验室, 北京 101318
  • 收稿日期:2021-11-23 修回日期:2022-01-12 接受日期:2022-01-17 发布日期:2022-03-02 出版日期:2022-11-10
  • 通讯作者: 冯迪
  • 作者简介:韩佳良(1997—),男,北京人,博士研究生,CCF会员,主要研究方向:分布式机器学习、联邦学习、推荐系统
    韩宇栋(2000—),男,山东东营人,博士研究生,主要研究方向:分布式机器学习、Web系统
    刘譞哲(1980—),男,甘肃兰州人,副教授,博士,CCF会员,主要研究方向:服务计算、系统软件
    赵耀帅(1977—),男,山东嘉祥人,高级工程师,硕士,主要研究方向:大数据、人工智能
    冯迪(1981—),女,湖北潜江人,工程师,硕士,主要研究方向:民航旅客行为分析、数据分析。fengdi@travelsky.com.cn
  • 基金资助:
    北大百度基金资助项目(2020BD007)

Personalized recommendation service system based on cloud-client-convergence

Jialiang HAN1, Yudong HAN1, Xuanzhe LIU1, Yaoshuai ZHAO2,3, Di FENG2,3()   

  1. 1.Key Laboratory of High Confidence Software Technologies of Ministry of Education (Peking University),Beijing 100871,China
    2.TravelSky Technology Limited,Beijing 101318,China
    3.Key Laboratory of Intelligent Application Technology for Civil Aviation Passenger Services,Civil Aviation Administration of China,Beijing 101318,China
  • Received:2021-11-23 Revised:2022-01-12 Accepted:2022-01-17 Online:2022-03-02 Published:2022-11-10
  • Contact: Di FENG
  • About author:HAN Jialiang, born in 1997, Ph. D. candidate. His research interests include distributed machine learning, federated learning, recommender system.
    HAN Yudong, born in 2000, Ph. D. candidate. His research interests include distributed machine learning, Web system.
    LIU Xuanzhe, born in 1980, Ph. D., associate professor. His research interests include service computing, system software.
    ZHAO Yaoshuai, born in 1977, M. S., senior engineer. His research interests include big data, artificial intelligence.
    FENG Di, born in 1981, M. S., engineer. Her research interests include civil aviation passenger behavior analysis, data analysis.
  • Supported by:
    PKU-Baidu Fund(2020BD007)

摘要:

主流个性化推荐服务系统通常利用部署在云端的模型进行推荐,因此需要将用户交互行为等隐私数据上传到云端,这会造成隐私泄露的隐患。为了保护用户隐私,可以在客户端处理用户敏感数据,然而,客户端存在通信瓶颈和计算资源瓶颈。针对上述挑战,设计了一个基于云?端融合的个性化推荐服务系统。该系统将传统的云端推荐模型拆分成用户表征模型和排序模型,在云端预训练用户表征模型后,将其部署到客户端,排序模型则部署到云端;同时,采用小规模的循环神经网络(RNN)抽取用户交互日志中的时序信息来训练用户表征,并通过Lasso算法对用户表征进行压缩,从而在降低云端和客户端之间的通信量以及客户端的计算开销的同时防止推荐准确率的下跌。基于RecSys Challenge 2015数据集进行了实验,结果表明,所设计系统的推荐准确率和GRU4REC模型相当,而压缩后的用户表征体积仅为压缩前的34.8%,计算开销较低。

关键词: 个性化推荐服务系统, 云?端融合, 用户表征模型, 隐私保护, 循环神经网络

Abstract:

Mainstream personalized recommendation systems usually use models deployed in the cloud to perform recommendation, so the private data such as user interaction behaviors need to be uploaded to the cloud, which may cause potential risks of user privacy leakage. In order to protect user privacy, user-sensitive data can be processed on the client, however, there are communication bottleneck and computation resource bottleneck in clients. Aiming at the above challenges, a personalized recommendation service system based on cloud-client-convergence was proposed. In this system, the cloud-based recommendation model was divided into a user representation model and a sorting model. After being pre-trained on the cloud, the user representation model was deployed to the client, while the sorting model was deployed to the cloud. A small-scale Recurrent Neural Network (RNN) was used to model the user behavior characteristics by extracting temporal information from user interaction logs, and the Lasso (Least absolute shrinkage and selection operator) algorithm was used to compress user representations, thereby preventing a drop in recommendation accuracy while reducing the communication overhead between the cloud and the client as well as the computation overhead of the client. Experiments were conducted on RecSys Challenge 2015 dataset, and the results show that the recommendation accuracy of the proposed system is comparable to that of the GRU4REC model, while the volume of the compressed user representations is only 34.8% of that before compression, with a lower computational overhead.

Key words: personalized recommendation service system, cloud-client-convergence, user representation model, privacy-preserving, Recurrent Neural Network (RNN)

中图分类号: