Social Media Event Prediction using DNN with Feedback Mechanism

Ma, W; Hu, X; Chen, C; Wen, S; Choo, KKR; Xiang, Y

Social Media Event Prediction using DNN with Feedback Mechanism

Ma, W Hu, X Chen, C Wen, S Choo, KKR Xiang, Y

Permalink

Publisher:: Association for Computing Machinery (ACM)
Publication Type:: Journal Article
Citation:: ACM Transactions on Management Information Systems, 2022, 13, (3), pp. 1-24
Issue Date:: 2022-09-01

Closed Access

	Filename	Description	Size
	3522759.pdf	Published version	2.28 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Ma, W
dc.contributor.author	Hu, X
dc.contributor.author	Chen, C
dc.contributor.author	Wen, S
dc.contributor.author	Choo, KKR
dc.contributor.author	Xiang, Y
dc.date.accessioned	2023-07-09T21:50:18Z
dc.date.available	2023-07-09T21:50:18Z
dc.date.issued	2022-09-01
dc.identifier.citation	ACM Transactions on Management Information Systems, 2022, 13, (3), pp. 1-24
dc.identifier.issn	2158-656X
dc.identifier.issn	2158-6578
dc.identifier.uri	http://hdl.handle.net/10453/171380
dc.description.abstract	<jats:p> Online social networks (OSNs) are a rich source of information, and the data (including user-generated content) can be mined to facilitate real-world event prediction. However, the dynamic nature of OSNs and the fast-pace nature of social events or hot topics compound the challenge of event prediction. This is a key limitation in many existing approaches. For example, our evaluations of six baseline approaches (i.e., logistic regression latent Dirichlet allocation (LDA)-based logistic regression (LR), multi-task learning (MTL), long short-term memory (LSTM) and convolutional neural networks, and transformer-based model) on three datasets collected as part of this research (two from Twitter and one from a news collection site <jats:xref ref-type="fn"> <jats:sup>1</jats:sup> </jats:xref> ), reveal that the accuracy of these approaches is between 50% and 60%, and they are not capable of utilizing new events in event predictions. Hence, in this article, we develop a novel DNN-based framework (hereafter referred to as event prediction with feedback mechanism— <jats:monospace>EPFM</jats:monospace> . Specifically, EPFM makes use of a feedback mechanism based on emerging events detection to improve the performance of event prediction. The feedback mechanism ensembles three outlier detection processes and returns a list of new events. Some of the events will then be chosen by analysts to feed into the fine-tuning process to update the predictive model. To evaluate EPFM, we conduct a series of experiments on the same three datasets, whose findings show that EPFM achieves 80% accuracy in event detection and outperforms the six baseline approaches.We also validate EPFM’s capability of detecting new events by empirically analyzing the feedback mechanism under different thresholds. </jats:p>
dc.language	en
dc.publisher	Association for Computing Machinery (ACM)
dc.relation.ispartof	ACM Transactions on Management Information Systems
dc.relation.isbasedon	10.1145/3522759
dc.rights	info:eu-repo/semantics/closedAccess
dc.subject	0806 Information Systems, 0807 Library and Information Studies
dc.subject.classification	4609 Information systems
dc.title	Social Media Event Prediction using DNN with Feedback Mechanism
dc.type	Journal Article
utslib.citation.volume	13
utslib.for	0806 Information Systems
utslib.for	0807 Library and Information Studies
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology/School of Computer Science
utslib.copyright.status	closed_access	*
dc.date.updated	2023-07-09T21:50:17Z
pubs.issue	3
pubs.publication-status	Published
pubs.volume	13
utslib.citation.issue	3

Abstract:

Online social networks (OSNs) are a rich source of information, and the data (including user-generated content) can be mined to facilitate real-world event prediction. However, the dynamic nature of OSNs and the fast-pace nature of social events or hot topics compound the challenge of event prediction. This is a key limitation in many existing approaches. For example, our evaluations of six baseline approaches (i.e., logistic regression latent Dirichlet allocation (LDA)-based logistic regression (LR), multi-task learning (MTL), long short-term memory (LSTM) and convolutional neural networks, and transformer-based model) on three datasets collected as part of this research (two from Twitter and one from a news collection site 1 ), reveal that the accuracy of these approaches is between 50% and 60%, and they are not capable of utilizing new events in event predictions. Hence, in this article, we develop a novel DNN-based framework (hereafter referred to as event prediction with feedback mechanism— EPFM . Specifically, EPFM makes use of a feedback mechanism based on emerging events detection to improve the performance of event prediction. The feedback mechanism ensembles three outlier detection processes and returns a list of new events. Some of the events will then be chosen by analysts to feed into the fine-tuning process to update the predictive model. To evaluate EPFM, we conduct a series of experiments on the same three datasets, whose findings show that EPFM achieves 80% accuracy in event detection and outperforms the six baseline approaches.We also validate EPFM’s capability of detecting new events by empirically analyzing the feedback mechanism under different thresholds.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/171380