Diabetweets: Analysis of Tweets for Health-Related Information

Hamzah Osop; Rabiul Hasan; Chei Sian Lee; Chee Yong Neo; Chee Kim Foo; Ankit Saurabh

doi:10.1007/978-3-030-60703-6_65

Diabetweets: Analysis of Tweets for Health-Related Information

Hamzah Osop^*, Rabiul Hasan, Chei Sian Lee, Chee Yong Neo, Chee Kim Foo, Ankit Saurabh

^*Corresponding author for this work

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

3 Citations (Scopus)

Abstract

Significant growth in health information sharing through Twitter is making it a compelling source for health-related information. Recent health research studies show Twitter data has been used for disease surveillance, health promotion, sentiment analysis, and perhaps has potential for clinical decision support. However, identifying health-related tweets in these massive Twitter datasets is challenging. With the increasing global prevalence of diabetes, user-generated health content in Twitter can be useful. Therefore, this preliminary study aims to classify diabetes-related tweets into meaningful health-related categories. Using an ensemble of neural network and stochastic gradient descent classifiers, we classified 13,667 diabetes-related tweets into five clusters. About 25.7% of the tweets were clustered as health-related, where 9.3% were classified as Treatment & Medication, 9.9% as Preventive Measures and 6.5% as Symptoms & Causes. More than 70% were clustered as Others. Analysing hashtags of tweets clustered in each of the categories showed significant relevance to health-related information.

Original language	English
Title of host publication	HCI International 2020 – Late Breaking Posters - 22nd International Conference, HCII 2020, Proceedings
Editors	Constantine Stephanidis, Margherita Antona, Stavroula Ntoa
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	500-508
Number of pages	9
ISBN (Print)	9783030607029
DOIs	https://doi.org/10.1007/978-3-030-60703-6_65
Publication status	Published - 2020
Externally published	Yes
Event	22nd International Conference on Human-Computer Interaction, HCI International 2020 - Copenhagen, Denmark Duration: Jul 19 2020 → Jul 24 2020

Publication series

Name	Communications in Computer and Information Science
Volume	1294
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	22nd International Conference on Human-Computer Interaction, HCI International 2020
Country/Territory	Denmark
City	Copenhagen
Period	7/19/20 → 7/24/20

Bibliographical note

Publisher Copyright:
© 2020, Springer Nature Switzerland AG.

ASJC Scopus Subject Areas

General Computer Science
General Mathematics

Keywords

Decision support
Health information sharing
Machine learning

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1007/978-3-030-60703-6_65

Cite this

Osop, H., Hasan, R., Lee, C. S., Neo, C. Y., Foo, C. K., & Saurabh, A. (2020). Diabetweets: Analysis of Tweets for Health-Related Information. In C. Stephanidis, M. Antona, & S. Ntoa (Eds.), HCI International 2020 – Late Breaking Posters - 22nd International Conference, HCII 2020, Proceedings (pp. 500-508). (Communications in Computer and Information Science; Vol. 1294). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-60703-6_65

Osop, Hamzah ; Hasan, Rabiul ; Lee, Chei Sian et al. / Diabetweets : Analysis of Tweets for Health-Related Information. HCI International 2020 – Late Breaking Posters - 22nd International Conference, HCII 2020, Proceedings. editor / Constantine Stephanidis ; Margherita Antona ; Stavroula Ntoa. Springer Science and Business Media Deutschland GmbH, 2020. pp. 500-508 (Communications in Computer and Information Science).

@inproceedings{d052d929db714be4b67e68fb33bcc736,

title = "Diabetweets: Analysis of Tweets for Health-Related Information",

abstract = "Significant growth in health information sharing through Twitter is making it a compelling source for health-related information. Recent health research studies show Twitter data has been used for disease surveillance, health promotion, sentiment analysis, and perhaps has potential for clinical decision support. However, identifying health-related tweets in these massive Twitter datasets is challenging. With the increasing global prevalence of diabetes, user-generated health content in Twitter can be useful. Therefore, this preliminary study aims to classify diabetes-related tweets into meaningful health-related categories. Using an ensemble of neural network and stochastic gradient descent classifiers, we classified 13,667 diabetes-related tweets into five clusters. About 25.7% of the tweets were clustered as health-related, where 9.3% were classified as Treatment & Medication, 9.9% as Preventive Measures and 6.5% as Symptoms & Causes. More than 70% were clustered as Others. Analysing hashtags of tweets clustered in each of the categories showed significant relevance to health-related information.",

keywords = "Decision support, Health information sharing, Machine learning",

author = "Hamzah Osop and Rabiul Hasan and Lee, {Chei Sian} and Neo, {Chee Yong} and Foo, {Chee Kim} and Ankit Saurabh",

note = "Publisher Copyright: {\textcopyright} 2020, Springer Nature Switzerland AG.; 22nd International Conference on Human-Computer Interaction, HCI International 2020 ; Conference date: 19-07-2020 Through 24-07-2020",

year = "2020",

doi = "10.1007/978-3-030-60703-6_65",

language = "English",

isbn = "9783030607029",

series = "Communications in Computer and Information Science",

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

pages = "500--508",

editor = "Constantine Stephanidis and Margherita Antona and Stavroula Ntoa",

booktitle = "HCI International 2020 – Late Breaking Posters - 22nd International Conference, HCII 2020, Proceedings",

address = "Germany",

}

Osop, H, Hasan, R, Lee, CS, Neo, CY, Foo, CK & Saurabh, A 2020, Diabetweets: Analysis of Tweets for Health-Related Information. in C Stephanidis, M Antona & S Ntoa (eds), HCI International 2020 – Late Breaking Posters - 22nd International Conference, HCII 2020, Proceedings. Communications in Computer and Information Science, vol. 1294, Springer Science and Business Media Deutschland GmbH, pp. 500-508, 22nd International Conference on Human-Computer Interaction, HCI International 2020, Copenhagen, Denmark, 7/19/20. https://doi.org/10.1007/978-3-030-60703-6_65

Diabetweets: Analysis of Tweets for Health-Related Information. / Osop, Hamzah; Hasan, Rabiul; Lee, Chei Sian et al.
HCI International 2020 – Late Breaking Posters - 22nd International Conference, HCII 2020, Proceedings. ed. / Constantine Stephanidis; Margherita Antona; Stavroula Ntoa. Springer Science and Business Media Deutschland GmbH, 2020. p. 500-508 (Communications in Computer and Information Science; Vol. 1294).

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

TY - GEN

T1 - Diabetweets

T2 - 22nd International Conference on Human-Computer Interaction, HCI International 2020

AU - Osop, Hamzah

AU - Hasan, Rabiul

AU - Lee, Chei Sian

AU - Neo, Chee Yong

AU - Foo, Chee Kim

AU - Saurabh, Ankit

PY - 2020

Y1 - 2020

N2 - Significant growth in health information sharing through Twitter is making it a compelling source for health-related information. Recent health research studies show Twitter data has been used for disease surveillance, health promotion, sentiment analysis, and perhaps has potential for clinical decision support. However, identifying health-related tweets in these massive Twitter datasets is challenging. With the increasing global prevalence of diabetes, user-generated health content in Twitter can be useful. Therefore, this preliminary study aims to classify diabetes-related tweets into meaningful health-related categories. Using an ensemble of neural network and stochastic gradient descent classifiers, we classified 13,667 diabetes-related tweets into five clusters. About 25.7% of the tweets were clustered as health-related, where 9.3% were classified as Treatment & Medication, 9.9% as Preventive Measures and 6.5% as Symptoms & Causes. More than 70% were clustered as Others. Analysing hashtags of tweets clustered in each of the categories showed significant relevance to health-related information.

AB - Significant growth in health information sharing through Twitter is making it a compelling source for health-related information. Recent health research studies show Twitter data has been used for disease surveillance, health promotion, sentiment analysis, and perhaps has potential for clinical decision support. However, identifying health-related tweets in these massive Twitter datasets is challenging. With the increasing global prevalence of diabetes, user-generated health content in Twitter can be useful. Therefore, this preliminary study aims to classify diabetes-related tweets into meaningful health-related categories. Using an ensemble of neural network and stochastic gradient descent classifiers, we classified 13,667 diabetes-related tweets into five clusters. About 25.7% of the tweets were clustered as health-related, where 9.3% were classified as Treatment & Medication, 9.9% as Preventive Measures and 6.5% as Symptoms & Causes. More than 70% were clustered as Others. Analysing hashtags of tweets clustered in each of the categories showed significant relevance to health-related information.

KW - Decision support

KW - Health information sharing

KW - Machine learning

UR - http://www.scopus.com/inward/record.url?scp=85097214614&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85097214614&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-60703-6_65

DO - 10.1007/978-3-030-60703-6_65

M3 - Conference contribution

AN - SCOPUS:85097214614

SN - 9783030607029

T3 - Communications in Computer and Information Science

SP - 500

EP - 508

BT - HCI International 2020 – Late Breaking Posters - 22nd International Conference, HCII 2020, Proceedings

A2 - Stephanidis, Constantine

A2 - Antona, Margherita

A2 - Ntoa, Stavroula

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 19 July 2020 through 24 July 2020

ER -

Osop H, Hasan R, Lee CS, Neo CY, Foo CK, Saurabh A. Diabetweets: Analysis of Tweets for Health-Related Information. In Stephanidis C, Antona M, Ntoa S, editors, HCI International 2020 – Late Breaking Posters - 22nd International Conference, HCII 2020, Proceedings. Springer Science and Business Media Deutschland GmbH. 2020. p. 500-508. (Communications in Computer and Information Science). doi: 10.1007/978-3-030-60703-6_65

Diabetweets: Analysis of Tweets for Health-Related Information

Abstract

Publication series

Conference

Bibliographical note

ASJC Scopus Subject Areas

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this