Journal of Computer Applications ›› 2021, Vol. 41 ›› Issue (10): 2806-2812.DOI: 10.11772/j.issn.1001-9081.2020111892

Special Issue: 人工智能

• Artificial intelligence • Previous Articles     Next Articles

Social recommendation based on dynamic integration of social information

REN Kezhou, PENG Furong, GUO Xin, WANG Zhe, ZHANG Xiaojing   

  1. Institute of Big Data Science and Industry, Shanxi University, Taiyuan Shanxi 030006, China
  • Received:2020-12-03 Revised:2021-03-04 Online:2021-10-10 Published:2021-10-27
  • Supported by:
    This work is partially supported by the Youth Program of National Natural Science Foundation of China (61802238), the Shanxi Province Key Research and Development Program (Social Development Side) (201903D321039), the Shanxi Province Science and Technology Innovation Project for Colleges and Universities (201802013), the Youth Science and Technology Research Fund of Shanxi Province Applied Basic Research Plan (201901D211168), the Shanxi Province Applied Basic Research Program (201901D111032).

动态融合社交信息的社会化推荐

任柯舟, 彭甫镕, 郭鑫, 王喆, 张晓静   

  1. 山西大学 大数据科学与产业研究院, 太原 030006
  • 通讯作者: 彭甫镕
  • 作者简介:任柯舟(1996-),男,山西稷山人,硕士研究生,CCF会员,主要研究方向:推荐系统;彭甫镕(1987-),男,贵州遵义人,副教授,博士,CCF会员,主要研究方向:数据挖掘、推荐系统;郭鑫(1982-),女,山西清徐人,讲师,博士,CCF会员,主要研究方向:文本挖掘;王喆(1993-),男,山西长治人,硕士研究生,主要研究方向:推荐系统;张晓静(1998-),女,山西原平人,CCF会员,主要研究方向:多源稀疏数据融合。
  • 基金资助:
    国家自然科学基金青年科学基金资助项目(61802238);山西省重点研发计划项目(社会发展方面)(201903D321039);山西省高等学校科技创新项目(201802013);山西省应用基础研究计划青年科技研究基金资助项目(201901D211168);山西省应用基础研究计划项目(201901D111032)。

Abstract: Aiming at the problem of data sparseness in recommendation algorithms, social data are usually introduced as auxiliary information for social recommendation. The traditional social recommendation algorithms ignore users' interest transfer, which makes the model unable to describe the dynamic characteristics of user interests, and the algorithms also ignore the dynamic characteristics of social influences, which causes the model to treat long before social behaviors and recent social behaviors equally. Aiming at these two problems, a social recommendation model named SLSRec with dynamic integration of social information was proposed. First, self-attention mechanism was used to construct a sequence model of user interaction items to implement the dynamic description of user interests. Then, an attention mechanism with forgetting with time was designed to model the short-term social interests, and an attention mechanism with collaborative characteristics was designed to model long-term social interests. Finally, the long-term and short-term social interests and the user's short-term interests were combined to obtain the user's final interests and generate the next recommendation. Normalized Discounted Cumulative Gain (NDCG) and Hit Rate (HR) indicators were used to compare and verify the proposed model, the sequence recommendation models (Self-Attention Sequence Recommendation (SASRec) model) and the social recommendation model (neural influence Diffusion Network for social recommendation (DiffNet) model) on the sparse dataset brightkite and the dense dataset Last.FM. Experimental results show that compared with DiffNet model, SLSRec model has the HR index increased by 8.5% on the sparse dataset; compared with SASRec model, SLSRec model has the NDCG index increased by 2.1% on the dense dataset, indicating that considering the dynamic characteristics of social information makes the recommendation results more accurate.

Key words: recommendation system, sequence recommendation, social recommendation, attention mechanism, information fusion

摘要: 针对推荐算法中的数据稀疏问题,通常引入社交数据作为辅助信息进行社会化推荐。传统的社会化推荐算法忽略用户的兴趣迁移,导致模型无法描述用户兴趣的动态变化特征,也忽略了社交影响的动态特性,导致模型将很久以前的社交行为与近期社交行为同等对待。针对这两点提出一种社交信息动态融合的社会化推荐模型SLSRec。首先,利用自注意力机制构建用户交互物品的序列模型,以实现对用户兴趣的动态描述;然后,设计具有时间遗忘的注意力机制对社交短期兴趣进行建模,并设计具有协同特性的注意力机制对社交长期兴趣进行建模;最后,融合社交的长短期兴趣与用户的短期兴趣来获得用户的最终兴趣并产生下一项推荐。利用归一化折损累计增益(NDCG)和命中率(HR)指标在稀疏数据集brightkite和稠密数据集Last.FM上把所提模型与序列推荐模型(自注意力序列推荐(SASRec)模型)和社会化推荐模型(社会推荐的神经影响扩散(DiffNet)模型)进行对比验证。实验结果显示,SLSRec模型与DiffNet模型相比,在稀疏数据集上的HR指标提升了8.5%;与SASRec模型相比,在稠密数据集上的NDCG指标提升了2.1%,表明考虑社交信息的动态特性使推荐结果更加准确。

关键词: 推荐系统, 序列推荐, 社会化推荐, 注意力机制, 信息融合

CLC Number: