计算机工程 ›› 2019, Vol. 45 ›› Issue (8): 203-209.doi: 10.19678/j.issn.1000-3428.0051987

• 人工智能及识别技术 • 上一篇    下一篇

基于时间效应的兴趣点推荐混合模型

张岐山1, 李可1, 林小榕2   

  1. 1. 福州大学 经济与管理学院, 福州 350108;
    2. 北京交通大学 下一代互联网互联设备国家工程实验室, 北京 100044
  • 收稿日期:2018-07-02 修回日期:2018-08-10 出版日期:2019-08-15 发布日期:2019-08-08
  • 作者简介:张岐山(1962-),男,教授、博士、博士生导师,主研方向为数据挖掘、推荐系统、系统优化与仿真;李可、林小榕,硕士研究生。
  • 基金项目:
    国家自然科学基金(61300104);福建省自然科学基金(2018J01791)。

Hybrid Model for Point-of-Interests Recommendation Based on Time Effect

ZHANG Qishan1, LI Ke1, LIN Xiaorong2   

  1. 1. School of Economics and Management, Fuzhou University, Fuzhou 350108, China;
    2. National Engineering Laboratory for Next Generation Internet Interconnection Devices, Beijing Jiaotong University, Beijing 100044, China
  • Received:2018-07-02 Revised:2018-08-10 Online:2019-08-15 Published:2019-08-08

摘要: 在基于位置的社交网络中,兴趣点实时推荐数据和用户签到数据存在高稀疏性问题。提出一种基于时间效应的混合推荐模型。通过用户潜在兴趣点数据模型计算用户时间行为影响分数和地理位置影响分数,并用线性统一模型进行处理,选取Top S个兴趣点作为用户的潜在兴趣点。将用户的潜在签到记录引入基于时间效应的矩阵分解模型中,考虑时间差异性和连续性对推荐结果的影响,在此基础上进行优化求解,提出推荐策略。实验结果表明,与LRT模型、UTE+SE模型相比,该模型的推荐效果较好,其准确率和召回率最高可达0.103 4和0.111 8。

关键词: 基于位置的社交网络, 时间信息, 地理位置信息, 矩阵填充, 矩阵分解, 实时推荐, 兴趣点

Abstract: In Location-Based Social Networks(LBSNs),real-time recommendation data of Point-of-Interests(POIs) and check-in data of users are highly sparse.Therefore,a hybrid recommendation model based on time effect is proposed.Through the data model of potential POIs of users,the time behavior influence scores and geographical location influence scores are calculated,and the linear unified model is used for processing.Top S POIs are selected as the potential POIs of users.Incorporate the user's potential check-in recorders into a time-based matrix factorization model,taking into account the influence of time difference and continuity on the recommendation results.On this basis,optimization solution is carried out,and the recommended strategy is proposed.Experimental results show that compared with the LRT model and UTE+SE model,the proposed model has better recommendation effect,and its precision and recall rate can reach up to 0.103 4 and 0.111 8.

Key words: Location-Based Social Networks(LBSNs), time information, geographical location information, matrix filling, matrix factorization, real-time recommendation, Point-of-Interests(POIs)

中图分类号: