计算机应用 ›› 2021, Vol. 41 ›› Issue (1): 60-66.DOI: 10.11772/j.issn.1001-9081.2020061010

所属专题: 第八届中国数据挖掘会议(CCDM 2020)

• 第八届中国数据挖掘会议(CCDM 2020) • 上一篇    下一篇

面向群组用户时序行为的动态推荐算法

温雯1, 刘芳1, 蔡瑞初1, 郝志峰1,2   

  1. 1. 广东工业大学 计算机学院, 广州 510000;
    2. 佛山科学技术学院 数学与大数据学院, 广东 佛山 528000
  • 收稿日期:2020-07-11 修回日期:2020-07-23 出版日期:2021-01-10 发布日期:2020-09-11
  • 通讯作者: 刘芳
  • 作者简介:温雯(1981-),女,江西赣州人,教授,博士,CCF会员,主要研究方向:推荐系统、数据挖掘、图表示学习;刘芳(1995-),女,江西吉安人,硕士研究生,CCF会员,主要研究方向:时序推荐、数据挖掘;蔡瑞初(1983-),男,浙江温州人,教授,博士,主要研究方向:因果关系发现、强化学习;郝志峰(1968-),男,江苏苏州人,教授,博士,主要研究方向:机器学习、人工智能。
  • 基金资助:
    广东省科技计划项目(2019A141401006)。

Dynamic recommendation algorithm for group-users' temporal behaviors

WEN Wen1, LIU Fang1, CAI Ruichu1, HAO Zhifeng1,2   

  1. 1. School of Computers, Guangdong University of Technology, Guangzhou Guangdong 510000, China;
    2. School of Mathematics and Big Data, Foshan University, Foshan Guangdong 528000, China
  • Received:2020-07-11 Revised:2020-07-23 Online:2021-01-10 Published:2020-09-11
  • Supported by:
    This work is partially supported by the Science and Technology Program of Guangdong Province (2019A141401006).

摘要: 针对现实系统中用户偏好随时间动态变化且一个用户ID背后可能是一个家庭的多个成员在共用的问题,提出一种为这类隐含多个类型成员行为的群组用户解决其偏好随时间而变化的动态推荐算法。首先,假设用户的历史行为数据包括曝光数据和点击数据,并通过学习当前时刻下群组用户的各类型角色权重来判别当前成员角色;其次,根据曝光数据提出两种设计思路来构造流行度模型,并采用逆倾向评分加权方法来平衡训练数据;最后,利用矩阵分解技术得出随时间变化的用户潜在偏好因子和物品潜在属性因子,计算两者内积后得出用户随时间变化的Top-K偏好推荐。实验结果表明,该算法在召回率、平均精度均值(MAP)、归一化折损累计增益(NDCG)这三个指标上一天24个时刻中均能有至少16个时刻的表现优于基准方法,并能缩短运行时间,降低计算的时间复杂度

关键词: 群组用户, 曝光数据, 时序行为, 矩阵分解, 偏好推荐

Abstract: Focusing on the issue that the user preferences change with time in the real system, and a user ID may be shared by multiple members of a family, a dynamic recommendation algorithm for the group-users who contained multiple types of members and have preferences varying with time was proposed. Firstly, it was assumed that the user's historical behavior data were composed of exposure data and click data, and the current member role was discriminated by learning the role weights of all types of members of the group-user at the present moment. Secondly, two design ideas were proposed according to the exposure data to construct a popularity model, and the training data were balanced by adopting the inverse propensity score weighting. Finally, the matrix factorization technique was used to obtain the user latent preference factor varying with time and the item latent attribute factor, and the inner products of the former and the latter were calculated to obtain the Top-K preference recommendations of the user which vary with time. Experimental results show that the proposed algorithm not only outperforms the benchmark method at least 16 moments in 24 moments a day on three metrics of Recall, Mean Average Precision (MAP), and Normalized Discounted Cumulative Gain (NDCG), but also shortens the running time and reduces the time complexity of calculation.

Key words: group-user, exposure data, temporal behavior, matrix factorization, preference recommendation

中图分类号: