LocBERT: Improving Social Media User Location Prediction Using Fine-Tuned BERT

Asif Khan; Huaping Zhang; Nada Boudjellal; Arshad Ahmad; Maqbool Khan

doi:10.1007/978-3-031-39689-2_3

LocBERT: Improving Social Media User Location Prediction Using Fine-Tuned BERT

Asif Khan, Huaping Zhang^*, Nada Boudjellal, Arshad Ahmad, Maqbool Khan

^*Corresponding author for this work

School of Computer Science and Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Predicting user locations on social media platforms like Twitter is a challenging task with numerous applications in marketing, politics, and disaster response. This paper introduces LocBERT, a fine-tuned BERT model designed to accurately predict the locations of Twitter users based on their conversations. Our experiments focus on the “StateElecTweets” dataset, comprising 1.6 million labeled tweets associated with state locations within the United States. The results demonstrate that LocBERT outperforms traditional machine learning models such as SVM and Naive Bayesian, achieving an accuracy of 0.988 and an F1-score of 0.987. The study contributes to predicting the location of users in election-related tweets, enabling a better understanding of campaign demographics and assisting political stakeholders in refining their strategies. The findings of this research hold significant implications for various domains and highlight the effectiveness of LocBERT in accurately predicting user locations on social media platforms.

Original language	English
Title of host publication	Database and Expert Systems Applications - DEXA 2023 Workshops - 34th International Conference, DEXA 2023, Proceedings
Editors	Gabriele Kotsis, Ismail Khalil, Atif Mashkoor, Johannes Sametinger, A Min Tjoa, Bernhard Moser, Maqbool Khan
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	23-32
Number of pages	10
ISBN (Print)	9783031396885
DOIs	https://doi.org/10.1007/978-3-031-39689-2_3
Publication status	Published - 2023
Event	34th International Conference on Database and Expert Systems Applications , DEXA 2023 - Penang, Malaysia Duration: 28 Aug 2023 → 30 Aug 2023

Publication series

Name	Communications in Computer and Information Science
Volume	1872 CCIS
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	34th International Conference on Database and Expert Systems Applications , DEXA 2023
Country/Territory	Malaysia
City	Penang
Period	28/08/23 → 30/08/23

Keywords

Location
Machine Learning
Social Media
Transformers

Access to Document

10.1007/978-3-031-39689-2_3

Cite this

Khan, A., Zhang, H., Boudjellal, N., Ahmad, A., & Khan, M. (2023). LocBERT: Improving Social Media User Location Prediction Using Fine-Tuned BERT. In G. Kotsis, I. Khalil, A. Mashkoor, J. Sametinger, A. M. Tjoa, B. Moser, & M. Khan (Eds.), Database and Expert Systems Applications - DEXA 2023 Workshops - 34th International Conference, DEXA 2023, Proceedings (pp. 23-32). (Communications in Computer and Information Science; Vol. 1872 CCIS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-39689-2_3

Khan, Asif ; Zhang, Huaping ; Boudjellal, Nada et al. / LocBERT : Improving Social Media User Location Prediction Using Fine-Tuned BERT. Database and Expert Systems Applications - DEXA 2023 Workshops - 34th International Conference, DEXA 2023, Proceedings. editor / Gabriele Kotsis ; Ismail Khalil ; Atif Mashkoor ; Johannes Sametinger ; A Min Tjoa ; Bernhard Moser ; Maqbool Khan. Springer Science and Business Media Deutschland GmbH, 2023. pp. 23-32 (Communications in Computer and Information Science).

@inproceedings{975fcdfc63ab4a5d9335e2f45619c2fe,

title = "LocBERT: Improving Social Media User Location Prediction Using Fine-Tuned BERT",

abstract = "Predicting user locations on social media platforms like Twitter is a challenging task with numerous applications in marketing, politics, and disaster response. This paper introduces LocBERT, a fine-tuned BERT model designed to accurately predict the locations of Twitter users based on their conversations. Our experiments focus on the “StateElecTweets” dataset, comprising 1.6 million labeled tweets associated with state locations within the United States. The results demonstrate that LocBERT outperforms traditional machine learning models such as SVM and Naive Bayesian, achieving an accuracy of 0.988 and an F1-score of 0.987. The study contributes to predicting the location of users in election-related tweets, enabling a better understanding of campaign demographics and assisting political stakeholders in refining their strategies. The findings of this research hold significant implications for various domains and highlight the effectiveness of LocBERT in accurately predicting user locations on social media platforms.",

keywords = "Location, Machine Learning, Social Media, Transformers",

author = "Asif Khan and Huaping Zhang and Nada Boudjellal and Arshad Ahmad and Maqbool Khan",

note = "Publisher Copyright: {\textcopyright} 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.; 34th International Conference on Database and Expert Systems Applications , DEXA 2023 ; Conference date: 28-08-2023 Through 30-08-2023",

year = "2023",

doi = "10.1007/978-3-031-39689-2_3",

language = "English",

isbn = "9783031396885",

series = "Communications in Computer and Information Science",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "23--32",

editor = "Gabriele Kotsis and Ismail Khalil and Atif Mashkoor and Johannes Sametinger and Tjoa, {A Min} and Bernhard Moser and Maqbool Khan",

booktitle = "Database and Expert Systems Applications - DEXA 2023 Workshops - 34th International Conference, DEXA 2023, Proceedings",

address = "Germany",

}

Khan, A, Zhang, H, Boudjellal, N, Ahmad, A & Khan, M 2023, LocBERT: Improving Social Media User Location Prediction Using Fine-Tuned BERT. in G Kotsis, I Khalil, A Mashkoor, J Sametinger, AM Tjoa, B Moser & M Khan (eds), Database and Expert Systems Applications - DEXA 2023 Workshops - 34th International Conference, DEXA 2023, Proceedings. Communications in Computer and Information Science, vol. 1872 CCIS, Springer Science and Business Media Deutschland GmbH, pp. 23-32, 34th International Conference on Database and Expert Systems Applications , DEXA 2023, Penang, Malaysia, 28/08/23. https://doi.org/10.1007/978-3-031-39689-2_3

LocBERT: Improving Social Media User Location Prediction Using Fine-Tuned BERT. / Khan, Asif; Zhang, Huaping; Boudjellal, Nada et al.
Database and Expert Systems Applications - DEXA 2023 Workshops - 34th International Conference, DEXA 2023, Proceedings. ed. / Gabriele Kotsis; Ismail Khalil; Atif Mashkoor; Johannes Sametinger; A Min Tjoa; Bernhard Moser; Maqbool Khan. Springer Science and Business Media Deutschland GmbH, 2023. p. 23-32 (Communications in Computer and Information Science; Vol. 1872 CCIS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - LocBERT

T2 - 34th International Conference on Database and Expert Systems Applications , DEXA 2023

AU - Khan, Asif

AU - Zhang, Huaping

AU - Boudjellal, Nada

AU - Ahmad, Arshad

AU - Khan, Maqbool

PY - 2023

Y1 - 2023

N2 - Predicting user locations on social media platforms like Twitter is a challenging task with numerous applications in marketing, politics, and disaster response. This paper introduces LocBERT, a fine-tuned BERT model designed to accurately predict the locations of Twitter users based on their conversations. Our experiments focus on the “StateElecTweets” dataset, comprising 1.6 million labeled tweets associated with state locations within the United States. The results demonstrate that LocBERT outperforms traditional machine learning models such as SVM and Naive Bayesian, achieving an accuracy of 0.988 and an F1-score of 0.987. The study contributes to predicting the location of users in election-related tweets, enabling a better understanding of campaign demographics and assisting political stakeholders in refining their strategies. The findings of this research hold significant implications for various domains and highlight the effectiveness of LocBERT in accurately predicting user locations on social media platforms.

AB - Predicting user locations on social media platforms like Twitter is a challenging task with numerous applications in marketing, politics, and disaster response. This paper introduces LocBERT, a fine-tuned BERT model designed to accurately predict the locations of Twitter users based on their conversations. Our experiments focus on the “StateElecTweets” dataset, comprising 1.6 million labeled tweets associated with state locations within the United States. The results demonstrate that LocBERT outperforms traditional machine learning models such as SVM and Naive Bayesian, achieving an accuracy of 0.988 and an F1-score of 0.987. The study contributes to predicting the location of users in election-related tweets, enabling a better understanding of campaign demographics and assisting political stakeholders in refining their strategies. The findings of this research hold significant implications for various domains and highlight the effectiveness of LocBERT in accurately predicting user locations on social media platforms.

KW - Location

KW - Machine Learning

KW - Social Media

KW - Transformers

UR - http://www.scopus.com/inward/record.url?scp=85171561702&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-39689-2_3

DO - 10.1007/978-3-031-39689-2_3

M3 - Conference contribution

AN - SCOPUS:85171561702

SN - 9783031396885

T3 - Communications in Computer and Information Science

SP - 23

EP - 32

BT - Database and Expert Systems Applications - DEXA 2023 Workshops - 34th International Conference, DEXA 2023, Proceedings

A2 - Kotsis, Gabriele

A2 - Khalil, Ismail

A2 - Mashkoor, Atif

A2 - Sametinger, Johannes

A2 - Tjoa, A Min

A2 - Moser, Bernhard

A2 - Khan, Maqbool

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 28 August 2023 through 30 August 2023

ER -

Khan A, Zhang H, Boudjellal N, Ahmad A, Khan M. LocBERT: Improving Social Media User Location Prediction Using Fine-Tuned BERT. In Kotsis G, Khalil I, Mashkoor A, Sametinger J, Tjoa AM, Moser B, Khan M, editors, Database and Expert Systems Applications - DEXA 2023 Workshops - 34th International Conference, DEXA 2023, Proceedings. Springer Science and Business Media Deutschland GmbH. 2023. p. 23-32. (Communications in Computer and Information Science). doi: 10.1007/978-3-031-39689-2_3

LocBERT: Improving Social Media User Location Prediction Using Fine-Tuned BERT

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this