On infectious intestinal disease surveillance using social media content

Bin Zou; Vasileios Lampos; Russell Gorton; Ingemar Johansson Cox

doi:10.1145/2896338.2896372

On infectious intestinal disease surveillance using social media content

Bin Zou, Vasileios Lampos, Russell Gorton, Ingemar Johansson Cox

Datalogisk Institut

21 Citationer (Scopus)

Abstract

This paper investigates whether infectious intestinal diseases (IIDs) can be detected and quantified using social media content. Experiments are conducted on user-generated data from the microblogging service, Twitter. Evaluation is based on the comparison with the number of IID cases reported by traditional health surveillance methods. We employ a deep learning approach for creating a topical vocabulary, and then apply a regularised linear (Elastic Net) as well as a nonlinear (Gaussian Process) regression function for inference. We show that like previous text regression tasks, the nonlinear approach performs better. In general, our experimental results, both in terms of predictive performance and semantic interpretation, indicate that Twitter data contain a signal that could be strong enough to complement conventional methods for IID surveillance.

Originalsprog	Engelsk
Titel	DH '16 : Proceedings of the 2016 Digital Health Conference
Antal sider	5
Forlag	Association for Computing Machinery
Publikationsdato	2016
Sider	157-161
ISBN (Elektronisk)	978-1-4503-4224-7
DOI	https://doi.org/10.1145/2896338.2896372
Status	Udgivet - 2016
Begivenhed	6th International Conference on Digital Health - Montreal, Canada Varighed: 11 apr. 2016 → 13 apr. 2016 Konferencens nummer: 6

Konference

Konference	6th International Conference on Digital Health
Nummer	6
Land/Område	Canada
By	Montreal
Periode	11/04/2016 → 13/04/2016

Adgang til dokumentet

10.1145/2896338.2896372

Citationsformater

@inproceedings{eba2f2b35483453ca87939b3d5b623a1,

title = "On infectious intestinal disease surveillance using social media content",

abstract = "This paper investigates whether infectious intestinal diseases (IIDs) can be detected and quantified using social media content. Experiments are conducted on user-generated data from the microblogging service, Twitter. Evaluation is based on the comparison with the number of IID cases reported by traditional health surveillance methods. We employ a deep learning approach for creating a topical vocabulary, and then apply a regularised linear (Elastic Net) as well as a nonlinear (Gaussian Process) regression function for inference. We show that like previous text regression tasks, the nonlinear approach performs better. In general, our experimental results, both in terms of predictive performance and semantic interpretation, indicate that Twitter data contain a signal that could be strong enough to complement conventional methods for IID surveillance.",

keywords = "Disease surveillance, IID, Infectious intestinal disease, Social media, Twitter, User-generated content, Word embeddings",

author = "Bin Zou and Vasileios Lampos and Russell Gorton and Cox, {Ingemar Johansson}",

year = "2016",

doi = "10.1145/2896338.2896372",

language = "English",

pages = "157--161",

booktitle = "DH '16",

publisher = "Association for Computing Machinery",

note = "6th International Conference on Digital Health, DH 2016 ; Conference date: 11-04-2016 Through 13-04-2016",

}

TY - GEN

T1 - On infectious intestinal disease surveillance using social media content

AU - Zou, Bin

AU - Lampos, Vasileios

AU - Gorton, Russell

AU - Cox, Ingemar Johansson

N1 - Conference code: 6

PY - 2016

Y1 - 2016

N2 - This paper investigates whether infectious intestinal diseases (IIDs) can be detected and quantified using social media content. Experiments are conducted on user-generated data from the microblogging service, Twitter. Evaluation is based on the comparison with the number of IID cases reported by traditional health surveillance methods. We employ a deep learning approach for creating a topical vocabulary, and then apply a regularised linear (Elastic Net) as well as a nonlinear (Gaussian Process) regression function for inference. We show that like previous text regression tasks, the nonlinear approach performs better. In general, our experimental results, both in terms of predictive performance and semantic interpretation, indicate that Twitter data contain a signal that could be strong enough to complement conventional methods for IID surveillance.

AB - This paper investigates whether infectious intestinal diseases (IIDs) can be detected and quantified using social media content. Experiments are conducted on user-generated data from the microblogging service, Twitter. Evaluation is based on the comparison with the number of IID cases reported by traditional health surveillance methods. We employ a deep learning approach for creating a topical vocabulary, and then apply a regularised linear (Elastic Net) as well as a nonlinear (Gaussian Process) regression function for inference. We show that like previous text regression tasks, the nonlinear approach performs better. In general, our experimental results, both in terms of predictive performance and semantic interpretation, indicate that Twitter data contain a signal that could be strong enough to complement conventional methods for IID surveillance.

KW - Disease surveillance

KW - IID

KW - Infectious intestinal disease

KW - Social media

KW - Twitter

KW - User-generated content

KW - Word embeddings

U2 - 10.1145/2896338.2896372

DO - 10.1145/2896338.2896372

M3 - Article in proceedings

AN - SCOPUS:84966605239

SP - 157

EP - 161

BT - DH '16

PB - Association for Computing Machinery

T2 - 6th International Conference on Digital Health

Y2 - 11 April 2016 through 13 April 2016

ER -

On infectious intestinal disease surveillance using social media content

Abstract

Konference

Adgang til dokumentet

Fingeraftryk

Citationsformater