AI helps you reading Science

AI generates interpretation videos

AI extracts and analyses the key points of the paper to generate videos automatically


pub
Go Generating

AI Traceability

AI parses the academic lineage of this thesis


Master Reading Tree
Generate MRT

AI Insight

AI extracts a summary of this paper


Weibo:
We have proposed LORE to incrementally mine sequential patterns from a check-in location stream of all users as a dynamic L2TG and derive a probability of a user visiting a new location based on the nth-order additive Markov chain

LORE: exploiting sequential influence for location recommendations

SIGSPATIAL/GIS, pp.103-112, (2014)

Cited by: 114|Views18
EI

Abstract

Providing location recommendations becomes an important feature for location-based social networks (LBSNs), since it helps users explore new places and makes LBSNs more prevalent to users. In LBSNs, geographical influence and social influence have been intensively used in location recommendations based on the facts that geographical proxi...More

Code:

Data:

0
Introduction
  • With the advancement of mobile devices and location acquisition technologies, location-based social networks (LBSNs) (e.g., Foursquare and Gowalla), have attracted millions of users [34].
  • In an LBSN (Figure 1), users can establish social links and share their experiences of visiting some specific locations, known as points-of-interest (POIs), e.g., restaurants, stores and museums.
  • These visits are known as check-in activities that reflect users’ preferences on locations.
  • People usually go to cinemas or bars after restaurants since they would like to relax after dinner
Highlights
  • With the advancement of mobile devices and location acquisition technologies, location-based social networks (LBSNs) (e.g., Foursquare and Gowalla), have attracted millions of users [34]
  • We propose a new LOcation REcommendation approach with sequential influence based on additive Markov chain (AMC), called LORE. (1) In LORE, we first incrementally mine sequential patterns from check-in location sequences of all users as a dynamic location-location transition graph (L2TG), where a location sequence consists of check-in locations of the same user ordered by check-in time
  • We have explored the sequential influence on users’ check-in behaviors in LBSNs
  • We have proposed LORE to incrementally mine sequential patterns from a check-in location stream of all users as a dynamic L2TG and derive a probability of a user visiting a new location based on the nth-order additive Markov chain (AMC)
  • We have integrated the sequential influence with the geographical influence and social influence into a unified location recommendation framework, in which the geographical influence is modeled as two-dimensional check-in distributions on locations instead of using one-dimensional distance distributions
  • Experimental results show that LORE provides much better location recommendations than all other recommendation techniques evaluated in our experiments
Results
  • The authors first compare the LORE against the state-of-the-art location recommendation techniques in terms of the recommendation accuracy in Section 6.1.
  • The authors present the performance comparison among the evaluated location recommendation techniques with regard to a large range of top-k values in Figure 4 and given-n values in Figure 5.
  • The authors conclude the most important and general findings in the experimental results on the two large-scale real data sets collected from Foursquare and Gowalla as follows.
  • FMC returns the most inaccurate POIs in terms of precision and misses most POIs visited by target users in terms of recall, as depicted in Figures 4 and 5
Conclusion
  • Discussion on Performance

    Which influence is better? According to Figures 4 and 5, AMC with the sequential influence is inferior to GS2D with the geographicalsocial influence on the Foursquare data set.
  • The main reason is that: AMC mines sequential patterns from all users’ check-in location sequences as a common location-location transition graph (L2TG) whereas GS2D separately models a two-dimensional check-in distribution for each user using her own check-in locations.
  • Both sequential and geographical-social influences are not conflicting to each other and can be integrated into a unified recommendation framework.
  • Experimental results show that LORE provides much better location recommendations than all other recommendation techniques evaluated in the experiments
Summary
  • Introduction:

    With the advancement of mobile devices and location acquisition technologies, location-based social networks (LBSNs) (e.g., Foursquare and Gowalla), have attracted millions of users [34].
  • In an LBSN (Figure 1), users can establish social links and share their experiences of visiting some specific locations, known as points-of-interest (POIs), e.g., restaurants, stores and museums.
  • These visits are known as check-in activities that reflect users’ preferences on locations.
  • People usually go to cinemas or bars after restaurants since they would like to relax after dinner
  • Results:

    The authors first compare the LORE against the state-of-the-art location recommendation techniques in terms of the recommendation accuracy in Section 6.1.
  • The authors present the performance comparison among the evaluated location recommendation techniques with regard to a large range of top-k values in Figure 4 and given-n values in Figure 5.
  • The authors conclude the most important and general findings in the experimental results on the two large-scale real data sets collected from Foursquare and Gowalla as follows.
  • FMC returns the most inaccurate POIs in terms of precision and misses most POIs visited by target users in terms of recall, as depicted in Figures 4 and 5
  • Conclusion:

    Discussion on Performance

    Which influence is better? According to Figures 4 and 5, AMC with the sequential influence is inferior to GS2D with the geographicalsocial influence on the Foursquare data set.
  • The main reason is that: AMC mines sequential patterns from all users’ check-in location sequences as a common location-location transition graph (L2TG) whereas GS2D separately models a two-dimensional check-in distribution for each user using her own check-in locations.
  • Both sequential and geographical-social influences are not conflicting to each other and can be integrated into a unified recommendation framework.
  • Experimental results show that LORE provides much better location recommendations than all other recommendation techniques evaluated in the experiments
Tables
  • Table1: Statistics of the two data sets Foursquare Gowalla
Download tables as Excel
Related work
  • In this section, we highlight related work on location recommendations. In general, there are four main categories for existing location recommendation approaches: collaborative filtering, social influence, geographical influence, and sequential influence.

    Collaborative filtering techniques. Some studies provide POI recommendations by using the conventional collaborative filtering techniques on users’ check-in data in LBSNs [14, 17, 23], GPS trajectory data [13, 35, 33], or text data [10]. However, the performance of all these techniques is considerably limited due to no consideration for the social influence, geographical influence, or sequential influence.

    Social influence. Based on the fact that nearby friends are more likely to share common interests, social link information has been widely utilized to improve the quality of recommender systems in LBSNs [1, 4, 8, 11, 15, 22, 24, 28, 30]. Specifically, these works derive the similarities between users in terms of their social links or/and residence distances, and then integrate them into the collaborative filtering techniques.
Funding
  • Jia-Dong Zhang and Chi-Yin Chow were partially supported by Guangdong Natural Science Foundation of China under Grant S2013010012363 and a research grant (CityU Project No 9231131)
Reference
  • J. Bao, Y. Zheng, and M. F. Mokbel. Location-based and preference-aware recommendation using sparse geo-social networking data. In ACM SIGSPATIAL, 2012.
    Google ScholarLocate open access versionFindings
  • Z. Chen, H. T. Shen, and X. Zhou. Discovering popular routes from trajectories. In IEEE ICDE, 2011.
    Google ScholarLocate open access versionFindings
  • A.-J. Cheng, Y.-Y. Chen, Y.-T. Huang, W. H. Hsu, and H.-Y. M. Liao. Personalized travel recommendation by mining people attributes from community-contributed photos. In ACM MM, 2011.
    Google ScholarLocate open access versionFindings
  • C. Cheng, H. Yang, I. King, and M. R. Lyu. Fused matrix factorization with geographical and social influence in location-based social networks. In AAAI, 2012.
    Google ScholarLocate open access versionFindings
  • C. Cheng, H. Yang, M. R. Lyu, and I. King. Where you like to go next: Successive point-of-interest recommendation. In IJCAI, 2013.
    Google ScholarFindings
  • E. Cho, S. A. Myers, and J. Leskovec. Friendship and mobility: User movement in location-based social networks. In ACM KDD, 2011.
    Google ScholarLocate open access versionFindings
  • G. Ference, M. Ye, and W.-C. Lee. Location recommendation for out-of-town users in location-based social networks. In ACM CIKM, 2013.
    Google ScholarLocate open access versionFindings
  • H. Gao, J. Tang, and H. Liu. gSCorr: Modeling geo-social correlations for new check-ins on location-based social networks. In ACM CIKM, 2012.
    Google ScholarLocate open access versionFindings
  • M. C. Gonzalez, C. A. Hidalgo, and A.-L. Barabasi. Understanding individual human mobility patterns. Nature, 453(7196):779–782, 2008.
    Google ScholarLocate open access versionFindings
  • B. Hu and M. Ester. Spatial topic modeling in online social media for location recommendation. In ACM RecSys, 2013.
    Google ScholarLocate open access versionFindings
  • T. Kurashima, T. Iwata, T. Hoshide, N. Takaya, and K. Fujimura. Geo topic model: Joint modeling of user’s activity area and interests for location recommendation. In ACM WSDM, 2013.
    Google ScholarLocate open access versionFindings
  • T. Kurashima, T. Iwata, G. Irie, and K. Fujimura. Travel route recommendation using geotags in photo sharing sites. In ACM CIKM, 2010.
    Google ScholarLocate open access versionFindings
  • K. W.-T. Leung, D. L. Lee, and W.-C. Lee. CLR: A collaborative location recommendation framework based on co-clustering. In ACM SIGIR, 2011.
    Google ScholarLocate open access versionFindings
  • J. J. Levandoski, M. Sarwat, A. Eldawy, and M. F. Mokbel. LARS: A location-aware recommender system. In IEEE ICDE, 2012.
    Google ScholarLocate open access versionFindings
  • B. Liu, Y. Fu, Z. Yao, and H. Xiong. Learning geographical preferences for point-of-interest recommendation. In ACM KDD, 2013.
    Google ScholarLocate open access versionFindings
  • X. Liu, Y. Liu, K. Aberer, and C. Miao. Personalized point-of-interest recommendation by mining users’ preference transition. In ACM CIKM, 2013. ACM SIGSPATIAL, 2012.
    Google ScholarLocate open access versionFindings
  • [18] B. W. Silverman. Density estimation for statistics and data analysis. Chapman and Hall, 1986.
    Google ScholarFindings
  • [19] C. Song, Z. Qu, N. Blumm, and A.-L. Barabasi. Limits of predictability in human mobility. Science, 327(5968):1018–1021, 2010.
    Google ScholarLocate open access versionFindings
  • [20] O. V. Usatenko, S. S. Apostolov, Z. A. Mayzelis, and S. S. Melnik. Random finite-valued dynamical systems: Additive Markov chain approach. Cambridge Scientific Publishers, Cambridge, 2009.
    Google ScholarLocate open access versionFindings
  • [21] C. Wang, M. Ye, and W.-C. Lee. From face-to-face gathering to social structure. In ACM CIKM, 2012.
    Google ScholarLocate open access versionFindings
  • [22] H. Wang, M. Terrovitis, and N. Mamoulis. Location recommendation in location-based social networks using user check-in data. In ACM SIGSPATIAL, 2013.
    Google ScholarLocate open access versionFindings
  • [23] M. Ye, P. Yin, and W.-C. Lee. Location recommendation for location-based social networks. In ACM SIGSPATIAL, 2010.
    Google ScholarLocate open access versionFindings
  • [24] M. Ye, P. Yin, W.-C. Lee, and D.-L. Lee. Exploiting geographical influence for collaborative point-of-interest recommendation. In ACM SIGIR, 2011.
    Google ScholarLocate open access versionFindings
  • [25] H. Yin, Y. Sun, B. Cui, Z. Hu, and L. Chen. LCARS: A location-content-aware recommender system. In ACM KDD, 2013.
    Google ScholarLocate open access versionFindings
  • [26] J. J.-C. Ying, W.-C. Lee, T.-C. Weng, and V. S. Tseng. Semantic trajectory mining for location prediction. In ACM SIGSPATIAL, 2011.
    Google ScholarLocate open access versionFindings
  • [27] Q. Yuan, G. Cong, Z. Ma, A. Sun, and N. Magnenat-Thalmann. Time-aware point-of-interest recommendation. In ACM SIGIR, 2013.
    Google ScholarLocate open access versionFindings
  • [28] J.-D. Zhang and C.-Y. Chow. iGSLR: Personalized geo-social location recommendation - a kernel density estimation approach. In ACM SIGSPATIAL, 2013.
    Google ScholarLocate open access versionFindings
  • [29] J.-D. Zhang and C.-Y. Chow. CoRe: Exploiting the personalized influence of two-dimensional geographic coordinates for location recommendations. Information Sciences, accepted to appear, 2014.
    Google ScholarLocate open access versionFindings
  • [30] J.-D. Zhang, C.-Y. Chow, and Y. Li. iGeoRec: A personalized and efficient geographical location recommendation framework. IEEE TSC, accepted to appear, 2014.
    Google ScholarFindings
  • [31] J.-D. Zhang, G. Ghinita, and C.-Y. Chow. Differentially private location recommendations in geosocial networks. In IEEE MDM, 2014.
    Google ScholarLocate open access versionFindings
  • [32] J.-D. Zhang, J. Xu, and S. S. Liao. Aggregating and sampling methods for processing GPS data streams for traffic state estimation. IEEE TITS, 14(4):1629–1641, 2013.
    Google ScholarLocate open access versionFindings
  • [33] V. W. Zheng, Y. Zheng, X. Xie, and Q. Yang. Towards mobile intelligence: Learning from GPS history data for collaborative recommendation. Artificial Intelligence, 184-185:17–37, 2012.
    Google ScholarLocate open access versionFindings
  • [34] Y. Zheng. Location-based social networks: Users. In Computing with Spatial Trajectories. Springer, 2011.
    Google ScholarFindings
  • [35] Y. Zheng and X. Xie. Learning location correlation from GPS trajectories. In IEEE MDM, 2010.
    Google ScholarLocate open access versionFindings
  • [36] Y.-T. Zheng, Z.-J. Zha, and T.-S. Chua. Mining travel patterns from geotagged photos. ACM TIST, 3(3):56:1–56:18, 2012.
    Google ScholarLocate open access versionFindings
Your rating :
0

 

Tags
Comments
数据免责声明
页面数据均来自互联网公开来源、合作出版商和通过AI技术自动分析结果,我们不对页面数据的有效性、准确性、正确性、可靠性、完整性和及时性做出任何承诺和保证。若有疑问,可以通过电子邮件方式联系我们:report@aminer.cn
小科