Journal of Computer Applications ›› 2019, Vol. 39 ›› Issue (5): 1261-1268.DOI: 10.11772/j.issn.1001-9081.2018102084

• Artificial intelligence • Previous Articles     Next Articles

Point-of-interest recommendation integrating social networks and image contents

SHAO Changcheng, CHEN Pinghua   

  1. School of Computers, Guangdong University of Technology, Guangzhou Guangdong 51000, China
  • Received:2018-10-15 Revised:2018-11-21 Online:2019-05-10 Published:2019-05-14
  • Supported by:
    This work is partially supported by the National Natural Science Foundation of China (61572144), the Science and Technology Planning Project of Guangdong Province (2016B030306002,2015B010110001, 2017B030307002).

融合社交网络和图像内容的兴趣点推荐

邵长城, 陈平华   

  1. 广东工业大学 计算机学院, 广州 510006
  • 通讯作者: 邵长城
  • 作者简介:邵长城(1992-),男,山东枣庄人,硕士研究生,CCF会员,主要研究方向:机器学习、大数据、推荐系统;陈平华(1967-),男,湖南攸县人,教授,硕士,主要研究方向:机器学习、大数据、推荐系统。
  • 基金资助:
    国家自然科学基金资助项目(61572144);广东省科技计划项目(2016B030306002,2015B010110001,2017B030307002)。

Abstract: The rapid growth of Location-Based Social Networks (LBSN) provides a vast amount of Point-of-Interest (POI) data, which facilitates the research of POI recommendation. To solve the low recommendation accuracy caused by the extreme sparseness of user-POI matrix and the lack of POI features, by integrating information such as tags, geography, socialization, score, and image information of POI, a POI recommendation method integrating social networks and image contents called SVPOI was proposed. Firstly, with the analysis of POI dataset, a distance factor was constructed based on power law distribution and a tag factor was constructed based on term frequency, and the existing historical score data was merged to construct a new user-POI matrix. Secondly, VGG16 Deep Convolutional Neural Network (DCNN) was used to process the images of POI to construct the POI image content matrix. Thirdly, the user social matrix was constructed according to the social network information of POI data. Finally, with the use of Probabilistic Matrix Factorization (PMF) model, the POI recommendation list was obtained with the integration of user-POI matrix, image content matrix and user social matrix. On real-world datasets, the accuracy of SVPOI is improved significantly compared to PMF, SoRec (Social Recommendation using probabilistic matrix factorization), TrustMF (Social Collaborative Filtering by Trust) and TrustSVD (Social Collaborative Filtering by Trust with SVD) while Mean Absolute Error (MAE) and Root-Mean-Square Error (RMSE) of SVPOI are decreased by 5.5% and 7.82% respectively compared to those of TrustMF which was the best of the comparison methods. The experimental results demonstrate the recommendation effectiveness of the proposed method.

Key words: point-of-interest recommendation, Location-Based Social Network (LBSN), image content, Deep Convolutional Neural Network (DCNN), Probabilistic Matrix Factorization (PMF) model

摘要: 基于位置的社交网络(LBSN)蓬勃发展,带来了大量的兴趣点(POI)数据,加速了兴趣点推荐的研究。针对用户-兴趣点矩阵极端稀疏造成的推荐精度低和兴趣点特征缺失问题,通过融合兴趣点的标签、地理、社交、评分以及图像等信息,提出了一种融合社交网络和图像内容的兴趣点推荐方法(SVPOI)。首先分析兴趣点数据集,针对地理信息,利用幂律概率分布构造距离因子;针对标签信息,利用检索词频率构造标签因子;融合已有的历史评分数据,构造新的用户-兴趣点评分矩阵。其次利用VGG16深度卷积神经网络模型(DCNN)识别兴趣点图像内容,构造兴趣点图像内容矩阵。然后根据兴趣点数据的社交网络信息,构造用户社交矩阵。最后,利用概率矩阵分解(PMF)模型,融合用户-兴趣点评分矩阵、图像内容矩阵、用户社交矩阵,构成SVPOI兴趣点推荐模型,生成兴趣点推荐列表。大量的真实数据集上的实验结果表明,与PMF、SoRec、TrustMF、TrustSVD推荐算法相比,SVPOI推荐的准确度均有较大提升,其平均绝对误差(MAE)和均方根误差(RMSE)两项指标比最优的TrustMF算法分别降低了5.5%和7.82%,表明SVPOI具有更好的推荐效果。

关键词: 兴趣点推荐, 基于位置的社交网络, 图像内容, 深度卷积神经网络, 概率矩阵分解模型

CLC Number: