Phillies tweeting from philly? Predicting twitter user locations with spatial word usage

Hau Wen Chang, Dongwon Lee, Mohammed Eltaher, Jeongkyu Lee

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

79 Citations (Scopus)

Abstract

We study the problem of predicting home locations of Twitter users using contents of their tweet messages. Using three probability models for locations, we compare both the Gaussian Mixture Model (GMM) and the Maximum Likelihood Estimation (MLE). In addition, we propose two novel unsupervised methods based on the notions of Non-Localness and Geometric-Localness to prune noisy data from tweet messages. In the experiments, our unsupervised approach improves the baselines significantly and shows comparable results with the supervised state-of-the-art method. For 5,113 Twitter users in the test set, on average, our approach with only 250 selected local words or less is able to predict their home locations (within 100 miles) with the accuracy of 0.499, or has 509.3 miles of average error distance at best.

Original languageEnglish (US)
Title of host publicationProceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012
Pages111-118
Number of pages8
DOIs
StatePublished - Dec 1 2012
Event2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012 - Istanbul, Turkey
Duration: Aug 26 2012Aug 29 2012

Publication series

NameProceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012

Other

Other2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012
CountryTurkey
CityIstanbul
Period8/26/128/29/12

Fingerprint

Maximum likelihood estimation
Experiments

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Software

Cite this

Chang, H. W., Lee, D., Eltaher, M., & Lee, J. (2012). Phillies tweeting from philly? Predicting twitter user locations with spatial word usage. In Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012 (pp. 111-118). [6425775] (Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012). https://doi.org/10.1109/ASONAM.2012.29
Chang, Hau Wen ; Lee, Dongwon ; Eltaher, Mohammed ; Lee, Jeongkyu. / Phillies tweeting from philly? Predicting twitter user locations with spatial word usage. Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012. 2012. pp. 111-118 (Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012).
@inproceedings{473b8cc392dd4e26a6591a4c02d59e9f,
title = "Phillies tweeting from philly? Predicting twitter user locations with spatial word usage",
abstract = "We study the problem of predicting home locations of Twitter users using contents of their tweet messages. Using three probability models for locations, we compare both the Gaussian Mixture Model (GMM) and the Maximum Likelihood Estimation (MLE). In addition, we propose two novel unsupervised methods based on the notions of Non-Localness and Geometric-Localness to prune noisy data from tweet messages. In the experiments, our unsupervised approach improves the baselines significantly and shows comparable results with the supervised state-of-the-art method. For 5,113 Twitter users in the test set, on average, our approach with only 250 selected local words or less is able to predict their home locations (within 100 miles) with the accuracy of 0.499, or has 509.3 miles of average error distance at best.",
author = "Chang, {Hau Wen} and Dongwon Lee and Mohammed Eltaher and Jeongkyu Lee",
year = "2012",
month = "12",
day = "1",
doi = "10.1109/ASONAM.2012.29",
language = "English (US)",
isbn = "9780769547992",
series = "Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012",
pages = "111--118",
booktitle = "Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012",

}

Chang, HW, Lee, D, Eltaher, M & Lee, J 2012, Phillies tweeting from philly? Predicting twitter user locations with spatial word usage. in Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012., 6425775, Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012, pp. 111-118, 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012, Istanbul, Turkey, 8/26/12. https://doi.org/10.1109/ASONAM.2012.29

Phillies tweeting from philly? Predicting twitter user locations with spatial word usage. / Chang, Hau Wen; Lee, Dongwon; Eltaher, Mohammed; Lee, Jeongkyu.

Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012. 2012. p. 111-118 6425775 (Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012).

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

TY - GEN

T1 - Phillies tweeting from philly? Predicting twitter user locations with spatial word usage

AU - Chang, Hau Wen

AU - Lee, Dongwon

AU - Eltaher, Mohammed

AU - Lee, Jeongkyu

PY - 2012/12/1

Y1 - 2012/12/1

N2 - We study the problem of predicting home locations of Twitter users using contents of their tweet messages. Using three probability models for locations, we compare both the Gaussian Mixture Model (GMM) and the Maximum Likelihood Estimation (MLE). In addition, we propose two novel unsupervised methods based on the notions of Non-Localness and Geometric-Localness to prune noisy data from tweet messages. In the experiments, our unsupervised approach improves the baselines significantly and shows comparable results with the supervised state-of-the-art method. For 5,113 Twitter users in the test set, on average, our approach with only 250 selected local words or less is able to predict their home locations (within 100 miles) with the accuracy of 0.499, or has 509.3 miles of average error distance at best.

AB - We study the problem of predicting home locations of Twitter users using contents of their tweet messages. Using three probability models for locations, we compare both the Gaussian Mixture Model (GMM) and the Maximum Likelihood Estimation (MLE). In addition, we propose two novel unsupervised methods based on the notions of Non-Localness and Geometric-Localness to prune noisy data from tweet messages. In the experiments, our unsupervised approach improves the baselines significantly and shows comparable results with the supervised state-of-the-art method. For 5,113 Twitter users in the test set, on average, our approach with only 250 selected local words or less is able to predict their home locations (within 100 miles) with the accuracy of 0.499, or has 509.3 miles of average error distance at best.

UR - http://www.scopus.com/inward/record.url?scp=84874276761&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84874276761&partnerID=8YFLogxK

U2 - 10.1109/ASONAM.2012.29

DO - 10.1109/ASONAM.2012.29

M3 - Conference contribution

AN - SCOPUS:84874276761

SN - 9780769547992

T3 - Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012

SP - 111

EP - 118

BT - Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012

ER -

Chang HW, Lee D, Eltaher M, Lee J. Phillies tweeting from philly? Predicting twitter user locations with spatial word usage. In Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012. 2012. p. 111-118. 6425775. (Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012). https://doi.org/10.1109/ASONAM.2012.29