Location prediction in large-scale social networks: an in-depth benchmarking study

Nur Al Hasan Haldar, Jianxin Li, Mark Reynolds, Timos Sellis, Jeffrey Xu Yu

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Location details of social users are important in diverse applications ranging from news recommendation systems to disaster management. However, user location is not easy to obtain from social networks because many users do not bother to provide this information or decline to do so due to privacy concerns. Thus, it is useful to estimate user locations from implicit information in the network. For this purpose, many location prediction models have been proposed that exploit different network features. Unfortunately, these models have not been benchmarked on common datasets using standard metrics. We fill this gap and provide an in-depth empirical comparison of eight representative prediction models using five metrics on four real-world large-scale datasets, namely Twitter, Gowalla, Brightkite, and Foursquare. We formulate a generalized procedure-oriented location prediction framework which allows us to evaluate and compare the prediction models systematically and thoroughly under extensive experimental settings. Based on our results, we perform a detailed analysis of the merits and limitations of the models providing significant insights into the location prediction problem.

Original languageEnglish
JournalVLDB Journal
DOIs
Publication statusE-pub ahead of print - 9 Jul 2019

Fingerprint

Benchmarking
Recommender systems
Disasters

Cite this

@article{765a0acfec4541e582d0237b202afaf5,
title = "Location prediction in large-scale social networks: an in-depth benchmarking study",
abstract = "Location details of social users are important in diverse applications ranging from news recommendation systems to disaster management. However, user location is not easy to obtain from social networks because many users do not bother to provide this information or decline to do so due to privacy concerns. Thus, it is useful to estimate user locations from implicit information in the network. For this purpose, many location prediction models have been proposed that exploit different network features. Unfortunately, these models have not been benchmarked on common datasets using standard metrics. We fill this gap and provide an in-depth empirical comparison of eight representative prediction models using five metrics on four real-world large-scale datasets, namely Twitter, Gowalla, Brightkite, and Foursquare. We formulate a generalized procedure-oriented location prediction framework which allows us to evaluate and compare the prediction models systematically and thoroughly under extensive experimental settings. Based on our results, we perform a detailed analysis of the merits and limitations of the models providing significant insights into the location prediction problem.",
keywords = "Experimental evaluation, Large social network, Location prediction",
author = "{Al Hasan Haldar}, Nur and Jianxin Li and Mark Reynolds and Timos Sellis and Yu, {Jeffrey Xu}",
year = "2019",
month = "7",
day = "9",
doi = "10.1007/s00778-019-00553-0",
language = "English",
journal = "The International Journal on Very Large Data Bases",
issn = "0949-877X",
publisher = "Springer",

}

Location prediction in large-scale social networks : an in-depth benchmarking study. / Al Hasan Haldar, Nur; Li, Jianxin; Reynolds, Mark; Sellis, Timos; Yu, Jeffrey Xu.

In: VLDB Journal, 09.07.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Location prediction in large-scale social networks

T2 - an in-depth benchmarking study

AU - Al Hasan Haldar, Nur

AU - Li, Jianxin

AU - Reynolds, Mark

AU - Sellis, Timos

AU - Yu, Jeffrey Xu

PY - 2019/7/9

Y1 - 2019/7/9

N2 - Location details of social users are important in diverse applications ranging from news recommendation systems to disaster management. However, user location is not easy to obtain from social networks because many users do not bother to provide this information or decline to do so due to privacy concerns. Thus, it is useful to estimate user locations from implicit information in the network. For this purpose, many location prediction models have been proposed that exploit different network features. Unfortunately, these models have not been benchmarked on common datasets using standard metrics. We fill this gap and provide an in-depth empirical comparison of eight representative prediction models using five metrics on four real-world large-scale datasets, namely Twitter, Gowalla, Brightkite, and Foursquare. We formulate a generalized procedure-oriented location prediction framework which allows us to evaluate and compare the prediction models systematically and thoroughly under extensive experimental settings. Based on our results, we perform a detailed analysis of the merits and limitations of the models providing significant insights into the location prediction problem.

AB - Location details of social users are important in diverse applications ranging from news recommendation systems to disaster management. However, user location is not easy to obtain from social networks because many users do not bother to provide this information or decline to do so due to privacy concerns. Thus, it is useful to estimate user locations from implicit information in the network. For this purpose, many location prediction models have been proposed that exploit different network features. Unfortunately, these models have not been benchmarked on common datasets using standard metrics. We fill this gap and provide an in-depth empirical comparison of eight representative prediction models using five metrics on four real-world large-scale datasets, namely Twitter, Gowalla, Brightkite, and Foursquare. We formulate a generalized procedure-oriented location prediction framework which allows us to evaluate and compare the prediction models systematically and thoroughly under extensive experimental settings. Based on our results, we perform a detailed analysis of the merits and limitations of the models providing significant insights into the location prediction problem.

KW - Experimental evaluation

KW - Large social network

KW - Location prediction

UR - http://www.scopus.com/inward/record.url?scp=85068818448&partnerID=8YFLogxK

U2 - 10.1007/s00778-019-00553-0

DO - 10.1007/s00778-019-00553-0

M3 - Article

JO - The International Journal on Very Large Data Bases

JF - The International Journal on Very Large Data Bases

SN - 0949-877X

ER -