Exploring temporal effects for location recommendation on location-based social networks

doi:10.1145/2507157.2507182

Home
/
Papers
/
Exploring temporal effects for location recommendation on location-based social networks

Proceedings Article•DOI•

Exploring temporal effects for location recommendation on location-based social networks

Huiji Gao¹, Jiliang Tang¹, Xia Hu¹, Huan Liu¹•Institutions (1)

Arizona State University¹

12 Oct 2013-pp 93-100

TL;DR: A novel location recommendation framework is introduced, based on the temporal properties of user movement observed from a real-world LBSN dataset, which exhibits the significance of temporal patterns in explaining user behavior, and demonstrates their power to improve location recommendation performance.

read less

Abstract: Location-based social networks (LBSNs) have attracted an inordinate number of users and greatly enriched the urban experience in recent years. The availability of spatial, temporal and social information in online LBSNs offers an unprecedented opportunity to study various aspects of human behavior, and enable a variety of location-based services such as location recommendation. Previous work studied spatial and social influences on location recommendation in LBSNs. Due to the strong correlations between a user's check-in time and the corresponding check-in location, recommender systems designed for location recommendation inevitably need to consider temporal effects. In this paper, we introduce a novel location recommendation framework, based on the temporal properties of user movement observed from a real-world LBSN dataset. The experimental results exhibit the significance of temporal patterns in explaining user behavior, and demonstrate their power to improve location recommendation performance.

...read moreread less

Citations

PDF

Open Access

More filters

Matrix Factorization Techniques for Recommender Systems

[...]

Patrick Seemann

01 Jan 2014

2,080 citations

Proceedings Article•DOI•

GeoMF: joint geographical modeling and matrix factorization for point-of-interest recommendation

[...]

Defu Lian¹, Cong Zhao¹, Xing Xie², Guangzhong Sun¹, Enhong Chen¹, Yong Rui² - Show less +2 more•Institutions (2)

University of Science and Technology of China¹, Microsoft²

24 Aug 2014

TL;DR: The results indicate that weighted matrix factorization is superior to other forms of factorization models and that incorporating the spatial clustering phenomenon in human mobility behavior on the LBSNs into matrixfactorization improves recommendation performance.

...read moreread less

Abstract: Point-of-Interest (POI) recommendation has become an important means to help people discover attractive locations However, extreme sparsity of user-POI matrices creates a severe challenge To cope with this challenge, viewing mobility records on location-based social networks (LBSNs) as implicit feedback for POI recommendation, we first propose to exploit weighted matrix factorization for this task since it usually serves collaborative filtering with implicit feedback better Besides, researchers have recently discovered a spatial clustering phenomenon in human mobility behavior on the LBSNs, ie, individual visiting locations tend to cluster together, and also demonstrated its effectiveness in POI recommendation, thus we incorporate it into the factorization model Particularly, we augment users' and POIs' latent factors in the factorization model with activity area vectors of users and influence area vectors of POIs, respectively Based on such an augmented model, we not only capture the spatial clustering phenomenon in terms of two-dimensional kernel density estimation, but we also explain why the introduction of such a phenomenon into matrix factorization helps to deal with the challenge from matrix sparsity We then evaluate the proposed algorithm on a large-scale LBSN dataset The results indicate that weighted matrix factorization is superior to other forms of factorization models and that incorporating the spatial clustering phenomenon into matrix factorization improves recommendation performance

...read moreread less

582 citations

Cites background or methods from "Exploring temporal effects for loca..."

..., Recall@k and Precision@k, in the top k POI recommendation [24, 12, 5]....
[...]
...Such a factorization algorithm has been exploited in [17, 5] for POI recommendation...
[...]
...For example, in [13, 25, 22], they tried to leverage content information of locations via topic modeling to assist POI recommendation; In [5], Gao et al....
[...]

Journal Article•DOI•

Recommendations in location-based social networks: a survey

[...]

Jie Bao¹, Yu Zheng², David Wilkie³, Mohamed F. Mokbel¹•Institutions (3)

University of Minnesota¹, Microsoft², University of North Carolina at Wilmington³

01 Jul 2015-Geoinformatica

TL;DR: A panorama of the recommender systems in location-based social networks with a balanced depth is presented, facilitating research into this important research theme.

...read moreread less

Abstract: Recent advances in localization techniques have fundamentally enhanced social networking services, allowing users to share their locations and location-related contents, such as geo-tagged photos and notes. We refer to these social networks as location-based social networks (LBSNs). Location data bridges the gap between the physical and digital worlds and enables a deeper understanding of users' preferences and behavior. This addition of vast geo-spatial datasets has stimulated research into novel recommender systems that seek to facilitate users' travels and social interactions. In this paper, we offer a systematic review of this research, summarizing the contributions of individual efforts and exploring their relations. We discuss the new properties and challenges that location brings to recommender systems for LBSNs. We present a comprehensive survey analyzing 1) the data source used, 2) the methodology employed to generate a recommendation, and 3) the objective of the recommendation. We propose three taxonomies that partition the recommender systems according to the properties listed above. First, we categorize the recommender systems by the objective of the recommendation, which can include locations, users, activities, or social media. Second, we categorize the recommender systems by the methodologies employed, including content-based, link analysis-based, and collaborative filtering-based methodologies. Third, we categorize the systems by the data sources used, including user profiles, user online histories, and user location histories. For each category, we summarize the goals and contributions of each system and highlight the representative research effort. Further, we provide comparative analysis of the recommender systems within each category. Finally, we discuss the available data-sets and the popular methods used to evaluate the performance of recommender systems. Finally, we point out promising research topics for future work. This article presents a panorama of the recommender systems in location-based social networks with a balanced depth, facilitating research into this important research theme.

...read moreread less

520 citations

Cites background or methods from "Exploring temporal effects for loca..."

...Foursquare 3 [33] Check-ins, Friendships, User Three sets of check-ins from 33,596 users...
[...]
...Foursquare 3 [33] This dataset7 contains three datasets from Foursquare: a) check-in history of 18107 users, b) check-in history of 11326 users, and c) 4163 users who live...
[...]
...[33] further extends the model with using more aggregated temporal functions, such as sum, mean, maximum and voting, over the users’ check-in data....
[...]

Proceedings Article•DOI•

Rank-GeoFM: A Ranking based Geographical Factorization Method for Point of Interest Recommendation

[...]

Xutao Li¹, Gao Cong¹, Xiaoli Li², Tuan-Anh Nguyen Pham¹, Shonali Krishnaswamy² - Show less +1 more•Institutions (2)

Nanyang Technological University¹, Institute for Infocomm Research Singapore²

09 Aug 2015

TL;DR: A ranking based geographical factorization method, called Rank-GeoFM, for POI recommendation, which addresses the two challenges of scarcity of check-in data and context information, and outperforms the state-of-the-art methods significantly in terms of recommendation accuracy.

...read moreread less

Abstract: With the rapid growth of location-based social networks, Point of Interest (POI) recommendation has become an important research problem. However, the scarcity of the check-in data, a type of implicit feedback data, poses a severe challenge for existing POI recommendation methods. Moreover, different types of context information about POIs are available and how to leverage them becomes another challenge. In this paper, we propose a ranking based geographical factorization method, called Rank-GeoFM, for POI recommendation, which addresses the two challenges. In the proposed model, we consider that the check-in frequency characterizes users' visiting preference and learn the factorization by ranking the POIs correctly. In our model, POIs both with and without check-ins will contribute to learning the ranking and thus the data sparsity problem can be alleviated. In addition, our model can easily incorporate different types of context information, such as the geographical influence and temporal influence. We propose a stochastic gradient descent based algorithm to learn the factorization. Experiments on publicly available datasets under both user-POI setting and user-time-POI setting have been conducted to test the effectiveness of the proposed method. Experimental results under both settings show that the proposed method outperforms the state-of-the-art methods significantly in terms of recommendation accuracy.

...read moreread less

340 citations

Cites background or methods from "Exploring temporal effects for loca..."

...• LRT: This is a recently developed matrix factorization method for POI recommendation with time information [7]....
[...]
...[7] study the temporal effect on the POI recommendation, but not the time-aware POI recommendation....
[...]
...By following previous studies [24, 7], we compute mtt∗ as:...
[...]
...Most of the existing POI recommendation methods [22, 3, 14, 9, 7, 12] overlook data scarcity and implicit feedback facts, and adapt conventional memory or model-based collaborative filtering for POI recommendation....
[...]

Proceedings Article•DOI•

Learning Graph-based POI Embedding for Location-based Recommendation

[...]

Min Xie¹, Hongzhi Yin², Hao Wang¹, Fanjiang Xu¹, Weitong Chen², Sen Wang³ - Show less +2 more•Institutions (3)

Chinese Academy of Sciences¹, University of Queensland², Griffith University³

24 Oct 2016

TL;DR: A generic graph-based embedding model is proposed that jointly captures the sequential effect, geographical influence, temporal cyclic effect and semantic effect in a unified way by embedding the four corresponding relational graphs into a shared low dimensional space and develops a novel time-decay method to dynamically compute the user's latest preferences.

...read moreread less

Abstract: With the rapid prevalence of smart mobile devices and the dramatic proliferation of location-based social networks (LBSNs), location-based recommendation has become an important means to help people discover attractive and interesting points of interest (POIs). However, the extreme sparsity of user-POI matrix and cold-start issue create severe challenges, causing CF-based methods to degrade significantly in their recommendation performance. Moreover, location-based recommendation requires spatiotemporal context awareness and dynamic tracking of the user's latest preferences in a real-time manner. To address these challenges, we stand on recent advances in embedding learning techniques and propose a generic graph-based embedding model, called GE, in this paper. GE jointly captures the sequential effect, geographical influence, temporal cyclic effect and semantic effect in a unified way by embedding the four corresponding relational graphs (POI-POI, POI-Region, POI-Time and POI-Word)into a shared low dimensional space. Then, to support the real-time recommendation, we develop a novel time-decay method to dynamically compute the user's latest preferences based on the embedding of his/her checked-in POIs learnt in the latent space. We conduct extensive experiments to evaluate the performance of our model on two real large-scale datasets, and the experimental results show its superiority over other competitors, especially in recommending cold-start POIs. Besides, we study the contribution of each factor to improve location-based recommendation and find that both sequential effect and temporal cyclic effect play more important roles than geographical influence and semantic effect.

...read moreread less

332 citations

Cites background from "Exploring temporal effects for loca..."

...As suggested in [7, 28], users’ mobility behaviors in the physical world exhibit strong temporal cyclic patterns, and the daily pattern (hours of the day) is one of the most fundamental patterns....
[...]
...By taking different definitions of temporal state, many other temporal patterns can be integrated into our GE model, as long as they contain the non-uniformness and consecutiveness properties [7]....
[...]
...[7] studied the temporal cyclic patterns of user check-ins in terms of temporal non-uniformness and temporal consecutiveness....
[...]
...Specifically, to overcome the data sparsity issue, most prior work of location-based recommendation focused on exploiting the geographical influence [10, 3, 20] and temporal cyclic effect [7, 28] to provide spatial or/and temporal context-aware recommendation....
[...]

1
2
3
4
…
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100

Collapse

References

PDF

Open Access

More filters

Journal Article•DOI•

Learning the parts of objects by non-negative matrix factorization

[...]

Daniel D. Lee¹, H. Sebastian Seung², H. Sebastian Seung¹•Institutions (2)

Alcatel-Lucent¹, Massachusetts Institute of Technology²

21 Oct 1999-Nature

TL;DR: An algorithm for non-negative matrix factorization is demonstrated that is able to learn parts of faces and semantic features of text and is in contrast to other methods that learn holistic, not parts-based, representations.

...read moreread less

Abstract: Is perception of the whole based on perception of its parts? There is psychological and physiological evidence for parts-based representations in the brain, and certain computational theories of object recognition rely on such representations. But little is known about how brains or computers might learn the parts of objects. Here we demonstrate an algorithm for non-negative matrix factorization that is able to learn parts of faces and semantic features of text. This is in contrast to other methods, such as principal components analysis and vector quantization, that learn holistic, not parts-based, representations. Non-negative matrix factorization is distinguished from the other methods by its use of non-negativity constraints. These constraints lead to a parts-based representation because they allow only additive, not subtractive, combinations. When non-negative matrix factorization is implemented as a neural network, parts-based representations emerge by virtue of two properties: the firing rates of neurons are never negative and synaptic strengths do not change sign.

...read moreread less

11,500 citations

"Exploring temporal effects for loca..." refers methods in this paper

...Furthermore, the better performance of LRT than NMF suggests that time-dependent check-in preference capture user mobile behavior better than static check-in preferences....
[...]
...• Non-negative Matrix Factorization (NMF) Non-negative Matrix Factorization (NMF) [13] computes nonnegative user check-in preferences under the whole user-location matrix, which is our basic location recommendation model, as defined in Eq....
[...]
...Both NMF and R-LRT perform better than CF, demonstrating their ability in dealing with sparse data for location recommendation....
[...]
...Non-negative Matrix Factorization (NMF) Non-negative Matrix Factorization (NMF) [13] computes nonnegative user check-in preferences under the whole user-location matrix, which is our basic location recommendation model, as de.ned in Eq....
[...]

Learning parts of objects by non-negative matrix factorization

[...]

D. D. Lee

01 Jan 1999

TL;DR: In this article, non-negative matrix factorization is used to learn parts of faces and semantic features of text, which is in contrast to principal components analysis and vector quantization that learn holistic, not parts-based, representations.

...read moreread less

9,604 citations

Journal Article•DOI•

Matrix Factorization Techniques for Recommender Systems

[...]

Yehuda Koren¹, Robert M. Bell, Chris Volinsky•Institutions (1)

Yahoo!¹

01 Aug 2009-IEEE Computer

TL;DR: As the Netflix Prize competition has demonstrated, matrix factorization models are superior to classic nearest neighbor techniques for producing product recommendations, allowing the incorporation of additional information such as implicit feedback, temporal effects, and confidence levels.

...read moreread less

Abstract: As the Netflix Prize competition has demonstrated, matrix factorization models are superior to classic nearest neighbor techniques for producing product recommendations, allowing the incorporation of additional information such as implicit feedback, temporal effects, and confidence levels

...read moreread less

9,583 citations

Proceedings Article•DOI•

Factorization meets the neighborhood: a multifaceted collaborative filtering model

[...]

Yehuda Koren¹•Institutions (1)

AT&T¹

24 Aug 2008

TL;DR: The factor and neighborhood models can now be smoothly merged, thereby building a more accurate combined model and a new evaluation metric is suggested, which highlights the differences among methods, based on their performance at a top-K recommendation task.

...read moreread less

Abstract: Recommender systems provide users with personalized suggestions for products or services. These systems often rely on Collaborating Filtering (CF), where past transactions are analyzed in order to establish connections between users and products. The two more successful approaches to CF are latent factor models, which directly profile both users and products, and neighborhood models, which analyze similarities between products or users. In this work we introduce some innovations to both approaches. The factor and neighborhood models can now be smoothly merged, thereby building a more accurate combined model. Further accuracy improvements are achieved by extending the models to exploit both explicit and implicit feedback by the users. The methods are tested on the Netflix data. Results are better than those previously published on that dataset. In addition, we suggest a new evaluation metric, which highlights the differences among methods, based on their performance at a top-K recommendation task.

...read moreread less

3,975 citations

"Exploring temporal effects for loca..." refers methods in this paper

...To solve largescale recommendation problems, matrix factorization is state-ofthe-art technology that has been proven to be successful in the Netflix Competition [11, 12], and is being used for item recommendation and trust prediction on product review sites like Epinions and Ciao for research purposes [15, 20, 21]....
[...]

Proceedings Article•DOI•

Friendship and mobility: user movement in location-based social networks

[...]

Eunjoon Cho¹, Seth A. Myers¹, Jure Leskovec¹•Institutions (1)

Stanford University¹

21 Aug 2011

TL;DR: A model of human mobility that combines periodic short range movements with travel due to the social network structure is developed and it is shown that this model reliably predicts the locations and dynamics of future human movement and gives an order of magnitude better performance.

...read moreread less

Abstract: Even though human movement and mobility patterns have a high degree of freedom and variation, they also exhibit structural patterns due to geographic and social constraints. Using cell phone location data, as well as data from two online location-based social networks, we aim to understand what basic laws govern human motion and dynamics. We find that humans experience a combination of periodic movement that is geographically limited and seemingly random jumps correlated with their social networks. Short-ranged travel is periodic both spatially and temporally and not effected by the social network structure, while long-distance travel is more influenced by social network ties. We show that social relationships can explain about 10% to 30% of all human movement, while periodic behavior explains 50% to 70%. Based on our findings, we develop a model of human mobility that combines periodic short range movements with travel due to the social network structure. We show that our model reliably predicts the locations and dynamics of future human movement and gives an order of magnitude better performance than present models of human mobility.

...read moreread less

2,922 citations

"Exploring temporal effects for loca..." refers background in this paper

...[5] proposed a Periodic & Social Mobility Model for location prediction with two temporal states (“home” and “work”) affected by social effects and non-social effects....
[...]
...Rm×n is a check-in indicator matrix, Yi j = 1 indicating correlations on LBSNs to solve the cold start location prediction that ui has checked in at lj, Yi j = 0otherwise. problem....
[...]
...Inspired by social influence theories that social friends tend to have similar check-in behavior, researches started to investigate the explicit social friendships on LBSNs [5, 9, 8] and leverage their power for improving location recommendation services [23, 3, 25]....
[...]
...However, these properties have not been exploited for location recommendation on LBSNs....
[...]
...Inspired by social in.uence theories that social friends tend to have similar check-in behavior, researches started to investigate the explicit social friendships on LBSNs [5, 9, 8] and leverage their power for improving location recommendation services [23, 3, 25]....
[...]