A point-of-interest suggestion algorithm in Multi-source geo-social networks☆
Introduction
Recent years have witnessed a blooming of Web 2.0, positioning systems and wireless communication technologies. Location based social networks (LBSNs), such as Foursquare2 and Yelp3 have become a popular application and facilitated users’ daily life. People can expand circles of friends, share their real experience and post reviews, photos and videos. As a significant tool of LBSNs, suggestion methods learn users’ preferences from their historical records and meta-data such as social relationship and review content, then suggest underlying points-of-interest (POIs) to the specific users. POI suggestion can benefit advertising agencies with an effective way of launching advertisements to the potential consumers, and improve user viscosity to LBSN service providers as well (Zhao et al., 2016a).
One of the troublesome problems in POI suggestion is data sparsity (Xie et al., 2016), especially in out-of-town areas. It stems from incompleteness of reviews posted in the LBSNs and degrades the accuracy of POI suggestion. When a user visits an area in his city or an out-of-town place which he is not familiar with, this problem becomes extremely severe and LBSNs can hardly suggest ideal results.
Meanwhile, the loose relationship of users in LBSNs make the suggestion results in out-of-town scenario trustless. Users are declined to trust friends in LBSNs since they usually have different backgrounds and interests. MBSNs (Microblog-Based Social Networks) (Xiong et al., 2018) such as Facebook and Twitter are platforms for users to communicate with close friends and share their current status. Sufficient valuable information is generated and can be used for POI suggestion. A typical application scenario is illustrated in Fig. 1. Due to the data sparsity of LBSNs, users are able to acquire extra information from Twitter, a MBSN, which can help a user make a more accurate decision. A geo-social network can be viewed as a heterogeneous information system (Bu et al., 2019, Cao et al., 2019, Li et al., 2019) composed of a LBSN and a MBSN.
Users’ preferences in LBSNs show some temporal and geographical patterns. POIs with similar attributes are usually clustered in a specific region. For instance, users prefer to visiting a street with more restaurants which can provide more choices, that is, a user’s visited places are often clustered together. A user is more likely to visit places nearby rather than the distant ones. Moreover, a user tends to visit a place at right time, e.g., he is more likely to visit a restaurant at lunch time rather than a cinema. Thus the spatio-temporal factors should be taken into account when suggesting POIs.
Inspired by these motivation, we introduce a latent probabilistic generative model called Multi-source Topic Awareness Suggestion (MTAS) to fuse the information from geo-social networks. Fig. 2 shows the basic idea of the proposed MTAS model. The geo-social network can be classified into four distinct information layers, and each layer corresponds to a part of “4W” (i.e., who, when, where and what) information structure. The content layer represents reviews from users; the social layer indicates different topologies of relationships between users in LBSNs and MBSNs, which generates different kinds of communities; the geographical layer and temporal layer show the spatio-temporal information of a posting event. The behavior of posting reviews about POIs bridges the gap between online social networks and offline physical world. Given a review, a layer link connects these four layers. An anchor link connects two different accounts of one user in a LBSN and a MBSN, respectively.
In order to study the applicability of MTAS, we explore its working mechanism and performance in two scenarios: (1) home-town suggestion is supposed that a specific user is located at his familiar regions such as his living or working areas. He appeals to suggestion services of LBSNs for new suggestion based on his activity records. (2) out-of-town suggestion satisfies the needs of users who travel to distant and unfamiliar areas. LBSNs can still suggest some POIs to the users but the results seem unreliable. It should be noticed that both the suggestion scenarios have its own spatio-temporal features.
In this study, we make the following original contributions:
(1) We model the multiple geo-social influence from social communities, time intervals and geographical locations to the textual content of reviews in both LBSNs and MBSNs based on the following facts: (a) users from a community might be interested in POIs located in the same region due to their frequent online communication; (b) the activity time of a community’s members tend to be tightly connected with each other; (c) a better POI suggestion result for a specific user can be inferred from review content extracted from LBSNs and MBSNs.
(2) We propose a latent probabilistic generative model called MTAS, which can accurately capture review words in LBSNs as well as MBSNs by taking into account information of social communities, time intervals and geographical locations. In addition, we can effectively infer the parameters of the MTAS model by the Gibbs sampling method.
(3) extensive experiments are conducted to evaluate the performance of the proposed MTAS model in two geo-social networks, and the experimental results show that our approach outperforms state-of-the-art baseline approaches in effectiveness and efficiency of POI suggestion.
The remainder of this paper is summarized as follows: Section 2 briefly reviews related works. Section 3 introduces some important definitions and formalize the POI suggestion problem. Section 4 describes the details of our proposed POI suggestion model, and then presents the inference process. Section 5 illustrates the application framework of POI suggestion. The experimental results are evaluated in Section 6. Lastly, we conclude this study in Section 7.
Section snippets
Related works
To improve the efficiency and efficacy of POI suggestion, some recent studies have attempted to examine and integrate textual content, spatio-temporal information as well as heterogeneous geo-social information.
Content effect. Textual content is an underlying factor which can reveal some significant information of POIs such as users’ sentiment and attitude (Xiong et al., 2019a, Xiong et al., 2019b). Ren et al. (2017) introduced a context-aware probabilistic matrix factorization approach for POI
Problem definition
In this section, we first introduce some notations of input data, and then present some significant definitions used in the study. Lastly, we briefly formalize the POI suggestion problem.
To facilitate understanding, Table 1 describes the important notations of input data appearing in this study.
Definition 1 POI A POI is a uniquely identified place (e.g., a cinema or a supermarket) or an event (e.g., a conference or a party) that someone finds useful or interesting. A POI has two attributes, i.e, identifier
Multi-source topic awareness suggestion model
In this section, we propose a Multi-source Topic AwarenessRecommendation model (MTAS model). Firstly, Section 4.1 introduces the model structure of MTAS. Then, Section 4.2 introduces the probabilistic generative process of observed variables, e.g., topics and review words of POIs, which are generated according to their distributions. Moreover, we infer the parameters in Sections 4.3 Model inference, 4.4 Inference framework. Lastly, we analyze the computational complexity of the inference
A POI suggestion framework via MTAS
In order to achieve our goal presented in Section 3, the POI suggestion framework is shown in Fig. 4.
Basically, the POI suggestion framework contains three essential phases: (1) Data is crawled from LBSN and MBSN portals and then we use the hybrid data to build geo-social networks. A series of geographical clusters called regions are generated by DBSCAN algorithm (Qiao et al., 2015a, Qiao et al., 2015b, Qiao et al., 2018a, Qiao et al., 2018b). Users are grouped into different communities by GN
Experimental results
A series of experiments were conducted to evaluate the effectiveness and efficiency of MTAS including the comparisons with the state-of-the-art methods in the environment with Java 8, Windows 10, and run on a PC with a CPU of Core i7 (7500U) and 8 GB RAM. The experimental results are presented in this section with the parameter 10.
Conclusion
POI Suggestion is a very challenging and difficult task in LBSNs. The data sparsity and user distrust in the LBSNs create a severe challenge for traditional recommender systems. Additionally, users’ behaviors in a LBSN show an obvious spatio-temporal pattern. In order to effectively recommend POIs in LBSNs, we integrate the factors including cross-platform textual content, temporal effect, social communities and geographical regions. Then, we propose a latent probabilistic generative model
Acknowledgments
This work was partially supported by the National Natural Science Foundation of China under Grant Nos. 61772091, 61802035, 61962006, 71701026, 61602064; the Sichuan Science and Technology Program under Grant Nos. 2018JY0448, 2019YFG0106, 2019YFS0067, 2018GZ0253, 2019YFS0236, 2017HH0088, 2018GZ0307; the Natural Science Foundation of Guangxi under Grant No. 2018GXNSFDA138005; the China Postdoctoral Science Foundation under Grant No. 2019M653400; the Youth Foundation for Humanities and Social
References (40)
- et al.
Effective successive POI recommendation inferred with individual behavior and group preference
Neurocomputing
(2016) - et al.
CSD: A multi-user similarity metric for community recommendation in online social networks
Expert Syst. Appl.
(2016) - et al.
Context-aware probabilistic matrix factorization modeling for point-of-interest recommendation
Neurocomputing
(2017) - et al.
An emotional contagion model for heterogeneous social media with multiple behaviors
Physica A
(2018) - et al.
Affective impression: Sentiment-awareness POI suggestion via embedding in heterogeneous LBSNs
IEEE Trans. Affect. Comput.
(2019) - et al.
How friends affect user behaviors? an exploration of social relation analysis for recommendation
Knowl.-Based Syst.
(2015) - et al.
A novel community detection algorithm based on simplification of complex networks
Knowl.-Based Syst.
(2017) - et al.
Latent dirichlet allocation
J. Mach. Learn. Res.
(2012) - et al.
Graph k-means based on leader identification, dynamic game and opinion dynamics
IEEE Trans. Knowl. Data Eng.
(2019) - et al.
Detecting prosumer-community groups in smart grids from the multiagent perspective
IEEE Trans. Syst. Man Cybern.: Syst.
(2019)
Cross-platform app recommendation by jointly modeling ratings and texts
ACM Trans. Inf. Syst.
Location recommendation for out-of-town users in location-based social networks
A personalized point-of-interest recommendation model via fusion of geo-social information
Neurocomputing
Spatial topic modeling in online social media for location recommendation
Topological influence-aware recommendation on social networks
Complexity
Who wants to join me?: Companion recommendation in location based social networks
Modeling human location data with mixtures of kernel densities
Unified point-of-interest recommendation with temporal interval assessment
Point-of-interest recommendation in location based social networks with topic and location awareness
A new modularity measure for Fuzzy community detection problems based on overlap and grouping functions
Internat. J. Approx. Reason.
Cited by (24)
NRDL: Decentralized user preference learning for privacy-preserving next POI recommendation
2024, Expert Systems with ApplicationsRandom walk-based algorithm for distance-aware influence maximization on multiple query locations
2022, Knowledge-Based SystemsThCoRe: Things of interest recommendation based on novel things correlations
2022, Information SciencesCitation Excerpt :The content-based method recommends things with an in-depth analysis on the ratings of users or things. The early work [11] proposes a user-based filtering method and a thing-based filtering method, where the former recommends things in terms of the similarity of user ratings on things, and the latter in terms of the similarity between things’ ratings. [12,13] expand the contents of things, associating some geographic information and social information with the contents, for recommendations.
Leveraging social influence based on users activity centers for point-of-interest recommendation
2022, Information Processing and ManagementCitation Excerpt :To this end, Li, Cong, Li, Pham, and Krishnaswamy (2015) proposed a geographical factorization model which considers the user’s check-in frequency and the location (i.e., latitude and longitude) of POIs to rank the POIs based on the user’s preferences. Inspired by the assumption which users who are friends have more common preferences than strangers users, several POI recommendation methods improve the performance of recommendation by taking social information into consideration (Gao, Li, Li, Song, & Zhou, 2018; Li, Wang, & Feng, 2020; Ma, King, & Lyu, 2009; Tang, Lin, Weng, & Zhu, 2016; Xiong et al., 2020). Zhang and Chow (2013) proposed a method called iGSLR that user preference, social influence, and personalized geographical influence are integrated into a unified geo-social recommendation framework.
A survey on deep learning based Point-of-Interest (POI) recommendations
2022, NeurocomputingCitation Excerpt :Earlier research works in POI recommendation primarily focused on feature engineering and conventional (non-deep learning) machine learning-based methods. Markov Chain based stochastic models have been explored extensively in this regard [1–9]. Due to the success of Matrix Factorization (MF [10]) based methods for recommendation systems in other domains, MF methods [2,11–20] have also been studied for better POI recommendation modeling.
- ☆
No author associated with this paper has disclosed any potential or pertinent conflicts which may be perceived to have impending conflict with this work. For full disclosure statements refer to https://doi.org/10.1016/j.engappai.2019.103374.
- 1
These authors contributed equally to this work.