LoCaTe: Influence Quantification for Location Promotion in Location-based Social Networks

Ankita Likhyani, Srikanta Bedathur, Deepak P.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Location-based social networks (LBSNs) such as Foursquare offer a platform for users to share and be aware of each other’s physical movements. Asa result of such a sharing of check-in information with each other, users can be influenced to visit (or check-in) at the locations visited by their friends. Quantifying such influences in these LBSNs is useful in various settings such as location promotion, personalized recommendations, mobility pattern prediction etc. In this paper, we focus on the problem of location promotion and develop a model to quantify the influence specific to a lo- cation between a pair of users. Specifically,we develop a joint model called LoCaTe, consisting of (i) user mobility model estimated using kernel density estimates; (ii) a model of the semantics of the location using topic models; and (iii) a modelof time-gap between check-ins using exponential distribution. We validate our model on a long term crawl of Foursquare data collected between Jan 2015 Feb 2016, as well as on publicly available LBSN datasets. Our experiments demonstrate that LoCaTe significantly outperforms state-of-theartmodels for the same task.
LanguageEnglish
Title of host publicationProceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17)
Pages2259-2265
Number of pages7
DOIs
Publication statusPublished - 05 Aug 2017
EventInternational Joint Conference on Artificial Intelligence (IJCAI 2017) - Melbourne, Melbourne, Australia
Duration: 19 Aug 201725 Aug 2017
https://ijcai-17.org/index.html

Conference

ConferenceInternational Joint Conference on Artificial Intelligence (IJCAI 2017)
Abbreviated titleIJCAI 2017
CountryAustralia
CityMelbourne
Period19/08/201725/08/2017
Internet address

Fingerprint

Semantics
Experiments

Cite this

Likhyani, A., Bedathur, S., & P., D. (2017). LoCaTe: Influence Quantification for Location Promotion in Location-based Social Networks. In Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17) (pp. 2259-2265) https://doi.org/10.24963/ijcai.2017/314
Likhyani, Ankita ; Bedathur, Srikanta ; P., Deepak. / LoCaTe: Influence Quantification for Location Promotion in Location-based Social Networks. Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17). 2017. pp. 2259-2265
@inproceedings{ff50315f012c4fb189d027c6ad40c0fe,
title = "LoCaTe: Influence Quantification for Location Promotion in Location-based Social Networks",
abstract = "Location-based social networks (LBSNs) such as Foursquare offer a platform for users to share and be aware of each other’s physical movements. Asa result of such a sharing of check-in information with each other, users can be influenced to visit (or check-in) at the locations visited by their friends. Quantifying such influences in these LBSNs is useful in various settings such as location promotion, personalized recommendations, mobility pattern prediction etc. In this paper, we focus on the problem of location promotion and develop a model to quantify the influence specific to a lo- cation between a pair of users. Specifically,we develop a joint model called LoCaTe, consisting of (i) user mobility model estimated using kernel density estimates; (ii) a model of the semantics of the location using topic models; and (iii) a modelof time-gap between check-ins using exponential distribution. We validate our model on a long term crawl of Foursquare data collected between Jan 2015 Feb 2016, as well as on publicly available LBSN datasets. Our experiments demonstrate that LoCaTe significantly outperforms state-of-theartmodels for the same task.",
author = "Ankita Likhyani and Srikanta Bedathur and Deepak P.",
year = "2017",
month = "8",
day = "5",
doi = "10.24963/ijcai.2017/314",
language = "English",
pages = "2259--2265",
booktitle = "Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17)",

}

Likhyani, A, Bedathur, S & P., D 2017, LoCaTe: Influence Quantification for Location Promotion in Location-based Social Networks. in Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17). pp. 2259-2265, International Joint Conference on Artificial Intelligence (IJCAI 2017), Melbourne, Australia, 19/08/2017. https://doi.org/10.24963/ijcai.2017/314

LoCaTe: Influence Quantification for Location Promotion in Location-based Social Networks. / Likhyani, Ankita; Bedathur, Srikanta; P., Deepak.

Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17). 2017. p. 2259-2265.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - LoCaTe: Influence Quantification for Location Promotion in Location-based Social Networks

AU - Likhyani, Ankita

AU - Bedathur, Srikanta

AU - P., Deepak

PY - 2017/8/5

Y1 - 2017/8/5

N2 - Location-based social networks (LBSNs) such as Foursquare offer a platform for users to share and be aware of each other’s physical movements. Asa result of such a sharing of check-in information with each other, users can be influenced to visit (or check-in) at the locations visited by their friends. Quantifying such influences in these LBSNs is useful in various settings such as location promotion, personalized recommendations, mobility pattern prediction etc. In this paper, we focus on the problem of location promotion and develop a model to quantify the influence specific to a lo- cation between a pair of users. Specifically,we develop a joint model called LoCaTe, consisting of (i) user mobility model estimated using kernel density estimates; (ii) a model of the semantics of the location using topic models; and (iii) a modelof time-gap between check-ins using exponential distribution. We validate our model on a long term crawl of Foursquare data collected between Jan 2015 Feb 2016, as well as on publicly available LBSN datasets. Our experiments demonstrate that LoCaTe significantly outperforms state-of-theartmodels for the same task.

AB - Location-based social networks (LBSNs) such as Foursquare offer a platform for users to share and be aware of each other’s physical movements. Asa result of such a sharing of check-in information with each other, users can be influenced to visit (or check-in) at the locations visited by their friends. Quantifying such influences in these LBSNs is useful in various settings such as location promotion, personalized recommendations, mobility pattern prediction etc. In this paper, we focus on the problem of location promotion and develop a model to quantify the influence specific to a lo- cation between a pair of users. Specifically,we develop a joint model called LoCaTe, consisting of (i) user mobility model estimated using kernel density estimates; (ii) a model of the semantics of the location using topic models; and (iii) a modelof time-gap between check-ins using exponential distribution. We validate our model on a long term crawl of Foursquare data collected between Jan 2015 Feb 2016, as well as on publicly available LBSN datasets. Our experiments demonstrate that LoCaTe significantly outperforms state-of-theartmodels for the same task.

U2 - 10.24963/ijcai.2017/314

DO - 10.24963/ijcai.2017/314

M3 - Conference contribution

SP - 2259

EP - 2265

BT - Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17)

ER -

Likhyani A, Bedathur S, P. D. LoCaTe: Influence Quantification for Location Promotion in Location-based Social Networks. In Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17). 2017. p. 2259-2265 https://doi.org/10.24963/ijcai.2017/314