Spatio-Temporal Recommendation in Social Media (eBook)

Dr. Hongzhi Yin has been an ARC DECRA fellow in the School of Information Technology and Electrical Engineering (ITEE), at The University of Queensland (UQ), and he received his PhD degree from Peking University in July 2014. His research interests include Recommender System and User Modeling, Social Media Mining and Management, Location-based Social Network Analysis, Deep Learning and Spatial Database.  Due to his great contributions to recommendation in social media, he was granted the Distinguished Doctor Degree Thesis Award of Peking University in 2014. Besides, he held the honors of outstanding graduate from Beijing provincial government of P.R. China. He was the winner of the National Scholarship from Ministry of Education of P.R. China in 2008 as well as the winner of National Graduate Scholarship from Ministry of Education of P.R. China in 2013. Dr. Yin has published over 30 related peer-reviewed publications in prestigious journals and conferences of the database, data mining and information retrieval fields, including SIGMOD, VLDB, KDD, ICDE, ACM Multimedia, CIKM, TOIS (ACM Transactions on Information Systems),  TKDD (ACM Transactions on Knowledge Discovery from Data), TIST (ACM Transactions on Intelligent Systems and Technology) and World Wide Web. He has served in the Technical Program Committee of various international conferences including IEEE International Conference on Data Science in Cyberspace 2016, WISE 2016&2015, APWEB 2016&2015, DEXA 2016&2015, WAIM 2016&2015. He has also serve as invited reviewers for several prestigious journals such as VLDB Journal, IEEE Transactions on Knowledge and Data Engineering (TKDE), ACM Transactions on the Web (TWeb), IEEE Transactions on Cybernetics, IEEE Transactions on Cloud Computing (TCC), Pervasive and Mobile Computing (PMC), New Review of Hypermedia and Multimedia, Frontiers of Computer Science (FCS), Journal of Image and Vision Computing, Knowledge-Based Systems, New Review of Hypermedia and Multimedia.Prof. Bin Cui is a faculty member at the School of EECS and Vice Director of the Institute of Network Computing and Information Systems, at Peking University. He obtained his BSc from Xi'an Jiaotong University (Pilot Class) in 1996, and his PhD from the National University of Singapore in 2004. From 2004 to 2006, he worked as a Research Fellow in the Singapore-MIT Alliance. His research interests include database system architectures, query and index techniques, and big data management and mining. He has served in the Technical Program Committee of various international conferences including SIGMOD, VLDB, ICDE and KDD, and as Vice PC Chair of ICDE 2011, Demo CO-Chair for ICDE 2014, and as Area Chair of VLDB 2014. He is currently on the Editorial Board of VLDB Journal, Distributed and Parallel Databases Journal, Information Systems, and Frontier of Computer Science, and was an associate editor of IEEE Transactions on Knowledge and Data Engineering (TKDE, 2009-2013). He has received the Microsoft Young Professorship award (MSRA 2008) and the CCF Young Scientist award (2009). He is a senior member of IEEE, member of ACM and distinguished member of CCF.

Preface 7
Acknowledgments 8
Contents 9
1 Introduction 12
1.1 Background 12
1.2 The Research Issues and Challenges 14
1.3 Overview of the Book 15
1.4 Literature and Research Review 17
1.4.1 Traditional Context-Aware Recommendation 17
1.4.2 Temporal Recommendation 18
1.4.3 Spatial Item Recommendation 19
1.4.4 Real-Time Recommendation 20
1.4.5 Online Recommendation Efficiency 21
References 23
2 Temporal Context-Aware Recommendation 27
2.1 Introduction 27
2.2 User Rating Behavior Modeling 30
2.2.1 Notations and Definitions 30
2.2.2 Temporal Context-Aware Mixture Model 32
2.2.3 Model Inference 34
2.2.4 Discussion About TCAM 36
2.2.5 Item-Weighting for TCAM 37
2.3 Temporal Recommendation 39
2.4 Experiments 40
2.4.1 Datasets 40
2.4.2 Comparisons 41
2.4.3 Evaluation Methodology 42
2.4.4 Recommendation Effectiveness 44
2.4.5 Temporal Context Influence Study 45
2.4.6 User Profile Analysis 47
2.5 Summary 48
References 48
3 Spatial Context-Aware Recommendation 50
3.1 Introduction 50
3.2 Location-Content-Aware Recommender System 53
3.2.1 Preliminary 53
3.2.2 Model Description 54
3.2.3 Model Inference 58
3.2.4 Online Recommendation 61
3.3 Experiments 61
3.3.1 Datasets 62
3.3.2 Comparative Approaches 63
3.3.3 Evaluation Methods 65
3.3.4 Recommendation Effectiveness 66
3.3.5 Local Preference Influence Study 68
3.3.6 Analysis of Latent Topic 70
3.4 Summary 71
References 71
4 Location-Based and Real-Time Recommendation 73
4.1 Introduction 74
4.1.1 Joint Modeling of User Check-In Behaviors 75
4.1.2 Real-Time POI Recommendation 76
4.2 Joint Modeling of User Check-In Activities 77
4.2.1 Preliminary 78
4.2.2 Model Structure 79
4.2.3 Generative Process 82
4.2.4 Model Inference 83
4.3 Online Learning for TRM 85
4.3.1 Feasibility Analysis 85
4.3.2 Online Learning Algorithm 86
4.4 POI Recommendation Using TRM 91
4.4.1 Fast Top-k Recommendation Framework 93
4.4.2 Addressing Cold-Start Problem 94
4.5 Experiments 94
4.5.1 Datasets 94
4.5.2 Comparative Approaches 96
4.5.3 Evaluation Methods 97
4.5.4 Recommendation Effectiveness 98
4.5.5 Impact of Different Factors 100
4.5.6 Test for Cold-Start Problem 102
4.5.7 Model Training Efficiency 103
4.6 Summary 104
References 104
5 Fast Online Recommendation 107
5.1 Introduction 107
5.1.1 Parallelization 108
5.1.2 Nearest-Neighbor Search 108
5.2 Metric Tree 110
5.2.1 Branch-and-Bound Algorithm 111
5.3 TA-Based Algorithm 113
5.3.1 Discussion 115
5.4 Attribute-Pruning Algorithm 116
5.5 Experiments 119
5.5.1 Experimental Results 119
5.6 Summary 121
References 122