ISSN 1000-1239 CN 11-1777/TP

计算机研究与发展 ›› 2020, Vol. 57 ›› Issue (3): 590-603.doi: 10.7544/issn1000-1239.2020.20190188

• 人工智能 • 上一篇    下一篇



  1. 1(电子科技大学信息与软件工程学院 成都 610054);2(提升政府治理能力大数据应用技术国家工程实验室(中电科大数据研究院有限公司) 贵阳 550022);3(中电科大数据研究院有限公司 贵阳 550022) (
  • 出版日期: 2020-03-01
  • 基金资助: 

Graph Embedding Based Session Perception Model for Next-Click Recommendation

Zeng Yifu1,2,3, Mu Qilin2,3, Zhou Le1, Lan Tian1, Liu Qiao1   

  1. 1(School of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu 610054);2(Big Data Application on Improving Government Governance Capabilities National Engineering Laboratory (CETC Big Data Research Institute Co., Ltd.), Guiyang 550022);3(CETC Big Data Research Institute Co., Ltd., Guiyang 550022)
  • Online: 2020-03-01
  • Supported by: 
    This work was supported by the National Science Foundation of China (U19B2028, 61772117), the Big Data Application on Improving Government Governance Capabilities National Engineering Laboratory Open Fund Project (10-2018039), the Sichuan Hi-Tech Industrialization Program (2018GFW0150), and the Fundamental Research Funds for the Central Universities (ZYGX2019J077).

摘要: 根据历史记录预测用户的下一次点击(即基于会话的推荐)是推荐系统中一个重要的子任务.重点研究会话推荐中如何在不牺牲预测准确性的情况下缓解用户的兴趣漂移问题,提高用户满意度.基本思想是从全局统计的角度出发,建立一个用于表示物品先后点击顺序的物品依赖关系图,据此提出一种图表示学习算法,生成可以保留关联物品间复杂关联关系的物品向量表达,最后,基于长/短期记忆机制,将物品向量表达作为“固定”输入,从而构建一个可以同时捕捉用户长期兴趣和短期兴趣的会话感知推荐模型.不同于其他相关工作,首次提出将下一次点击预测模型建立在“固定”物品表达的基础上.在公开数据集上的实验结果表明:提出的推荐模型在预测准确性和推荐多样新颖性上的表现优于其他相关方法.

关键词: 基于会话的推荐系统, 行为建模, 图表示学习, 用户兴趣, 神经网络

Abstract: Predicting users’ next-click according to their historical session records, also known as session-based recommendation, is an important and challenging task and has led to a considerable amount of work towards this aim. Several significant progresses have been made in this area, but some fundamental problems still remain open, such as the trade-off between users’ satisfaction and predictive accuracy of the models. In this study, we consider the problem of how to alleviate user interests drift without sacrificing the predictive accuracy. For this purpose, we first set up an item dependency graph to represent the click sequence of items from a global, statistical perspective. Then an efficient graph embedding learning algorithm is proposed to produce item embeddings which preserve the information flow properties of the system and the structural dependency between each pair of items. Finally, the proposed model is capable of capturing the users’ general interests and their temporal browsing interests simultaneously by using of a BiLSTM based long/short term memory mechanism. Experimental results on two real-world data sets show that the proposed model not only performs better in terms of predictive accuracy but also demonstrates better diversity and novelty in its recommendations as compared with other state-of-the-art methods.

Key words: session-based recommendation system, behavior modeling, graph representation learning, user interests, neural network