梁弼,刘笃晋,熊伦,等.一种内容和地点感知的个性化POI推荐模型[J].深圳大学学报理工版,2022,39(6):693-700.[doi:10.3724/SP.J.1249.2022.06693]
LIANG Bi,LIU Dujin,XIONG Lun,et al.A content-location-aware personalized POI recommendation model[J].Journal of Shenzhen University Science and Engineering,2022,39(6):693-700.[doi:10.3724/SP.J.1249.2022.06693]
点击复制
一种内容和地点感知的个性化POI推荐模型
梁弼1,2，刘笃晋1，熊伦1，许晓红1

1.四川文理学院智能制造学院，四川达州635000;2.北京邮电大学计算机学院，北京100876

人工智能;兴趣点推荐;数据稀疏性;内容主题;地点主题;上下文因素;潜在关系;概率生成模型;位置社交网络

A content-location-aware personalized POI recommendation model
LIANG Bi1,2，LIU Dujin1，XIONG Lun1，and XU Xiaohong1

1.School of Intelligent Manufacturing, Sichuan University of Arts and Science, Dazhou 635000, Sichuan Province, P. R. China;2.School of Computer Science, Beijing University of Posts and Telecommunications, Beijing 100876, P. R. China

artificial intelligence; point of interest recommendation; data sparsity; content topic; location topic; context factors; latent relation; probabilistic generative model; location-based social network

DOI: 10.3724/SP.J.1249.2022.06693

备注

针对兴趣点（pointofinterest，POI）推荐中用户-POI交互矩阵数据稀疏问题，当前研究仅通过探索地理位置、内容信息及社会关系等上下文因素来缓解该问题，缺乏对这些上下文因素共同作用情况的综合分析及利用．为此，采用概率生成的方法提出一种内容和地点感知的主题模型（content-location-awaretopicmodel，CLATM），用以模拟用户在决策过程中的签到行为．该模型由内容主题建模和地点主题建模两个核心模块构成，用户签到内容依赖内容主题和地点主题，内容主题和地点主题在一定程度上共同决定用户签到地点，地理位置依赖于地点主题并服从高斯分布．该模型不仅恰当地整合了内容、地点和地理位置等重要的上下文因素，且充分利用这些因素之间的潜在关系有效缓解了数据稀疏性．在Foursquare和Yelp两个真实的位置社交网络数据集上对CLATM进行性能评测，实验结果表明，该模型在召回率（recall）和归一化折损累计增益（normalizeddiscountedcumulativegain，NDCG）指标上均优于基准，recall@20和NDCG@20最大分别提高约141.09%和94.44%．综合使用上下文因素的共同作用能有效提升POI推荐性能．
Aiming at the data sparsity problem of user-POI matrix in point of interest (POI) recommendation, the more and more studies have explored the contextual factors such as geographical location, content information and social relations to deal with the above-mentioned problem. However, the current research lacks comprehensive analysis and utilization of the relations of these contextual factors. Therefore, we propose a content-location-aware topic model (CLATM) to simulate the user check-in behavior in the decision-making process from the dual-perspective of content and location by using the probability generation method. CLATM consists of two core modules: content topic modeling and location topic modeling. The user check-in content depends on the content topic and location topic. The content topic and location topic jointly determine the user check-in location to a certain extent. The geographic location depends on the location topic and obeys Gaussian distribution. The CLATM model not only properly integrates the important contextual factors such as content, location and geography, but also makes full use of the latent relations between these factors to alleviate the data sparsity effectively. The performance of CLATM model is evaluated on two real location-based social network (LBSN) datasets, Foursquare and Yelp. The experimental results show that the model is superior to the baselines in recall and normalize discounted cumulative gain (NDCG), with the maximum increase of about 141. 09% and 94. 44% in recall@20 and NDCG@20, respectively. It can be concluded that comprehensive use of the relations of contextual factors can effectively improve the POI recommendation performance.
·