Mining event-oriented topics in microblog stream with unsupervised multi-view hierarchical embedding

Peng, M, Zhu, J, Wang, H, Li, X, Zhang, Y, Zhang, X and Tian, G 2018, 'Mining event-oriented topics in microblog stream with unsupervised multi-view hierarchical embedding', ACM Transactions on Knowledge Discovery from Data, vol. 12, no. 3, pp. 1-26.


Document type: Journal Article
Collection: Journal Articles

Title Mining event-oriented topics in microblog stream with unsupervised multi-view hierarchical embedding
Author(s) Peng, M
Zhu, J
Wang, H
Li, X
Zhang, Y
Zhang, X
Tian, G
Year 2018
Journal name ACM Transactions on Knowledge Discovery from Data
Volume number 12
Issue number 3
Start page 1
End page 26
Total pages 26
Publisher Association for Computing Machinery
Abstract This article presents an unsupervised multi-view hierarchical embedding (UMHE) framework to sufficiently reveal the intrinsic topical knowledge in social events. Event-oriented topics are highly related to such events as it can provide explicit descriptions of what have happened in social community. In many real-world cases, however, it is difficult to include all attributes of microblogs, more often, textual aspects only are available. Traditional topic modelling methods have failed to generate event-oriented topics with the textual aspects, since the inherent relations between topics are often overlooked in these methods. Meanwhile, the metrics in original word vocabulary space might not effectively capture semantic distances. Our UMHE framework overcomes the severe information deficiency and poor feature representation. The UMHE first develops a multi-view Bayesian rose tree to preliminarily generate prior knowledge for latent topics and their relations. With such prior knowledge, we design an unsupervised translation-based hierarchical embedding method to make a better representation of these latent topics. By applying self-adaptive spectral clustering on the embedding space and the original space concomitantly, we eventually extract event-oriented topics in word distributions to express social events. Our framework is purely data-driven and unsupervised, without any external knowledge. Experimental results on TREC Tweets2011 dataset and Sina Weibo dataset demonstrate that the UMHE framework can construct hierarchical structure with high fitness, but also yield topic embeddings with salient semantics; therefore, it can derive event-oriented topics with meaningful descriptions.
Subject Natural Language Processing
Keyword(s) Bayesian rose tree
Event-oriented topic
Multi-view hierarchical embedding
Unsupervised learning
DOI - identifier 10.1145/3173044
Copyright notice © 2018 ACM
ISSN 1556-4681
Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 15 times in Thomson Reuters Web of Science Article | Citations
Scopus Citation Count Cited 0 times in Scopus Article
Altmetric details:
Access Statistics: 23 Abstract Views  -  Detailed Statistics
Created: Tue, 23 Oct 2018, 16:00:00 EST by Catalyst Administrator
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us