Current studentsExisting staffContact us
 

DATM: A Novel Data Agnostic Topic Modeling Technique With Improved Effectiveness for Both Short and Long Text

Simple item page
dc.contributor.authorBewong, M.
dc.contributor.authorWondoh, J.
dc.contributor.authorKwashie, S.
dc.contributor.authorLiu, J.
dc.contributor.authorLiu, L.
dc.contributor.authorLi, J.
dc.contributor.authorIslam, M.Z.
dc.contributor.authorKernot, D.
dc.date.issued2023
dc.description.abstractTopic modelling is important for tackling several data mining tasks in information retrieval. While seminal topic modelling techniques such as Latent Dirichlet Allocation (LDA) have been proposed, the ubiquity of social media and the brevity of its texts pose unique challenges for such traditional topic modelling techniques. Several extensions including <italic>auxiliary aggregation</italic>, <italic>self aggregation</italic> and <italic>direct learning</italic> have been proposed to mitigate these challenges, however some still remain. These include a lack of consistency in the topics generated and the decline in model performance in applications involving disparate document lengths. There is a recent paradigm shift towards neural topic models, which are not suited for resource-constrained environments. This paper revisits LDA-style techniques, taking a theoretical approach to analyse the relationship between word co-occurrence and topic models. Our analysis shows that by altering the word co-occurrences within the corpus, topic discovery can be enhanced. Thus we propose a novel data transformation approach dubbed <italic>DATM</italic> to improve the topic discovery within a corpus. A rigorous empirical evaluation shows that DATM is not only powerful, but it can also be used in conjunction with existing benchmark techniques to significantly improve their effectiveness and their consistency by up to 2 fold.
dc.identifier.citationIEEE Access, 2023; 11:32826-32841
dc.identifier.doi10.1109/ACCESS.2023.3262653
dc.identifier.issn2169-3536
dc.identifier.issn2169-3536
dc.identifier.orcidLiu, J. [0000-0002-0794-0404]
dc.identifier.urihttps://hdl.handle.net/11541.2/33676
dc.language.isoen
dc.publisherIEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
dc.source.urihttps://doi.org/10.1109/access.2023.3262653
dc.titleDATM: A Novel Data Agnostic Topic Modeling Technique With Improved Effectiveness for Both Short and Long Text
dc.typeJournal article
pubs.publication-statusPublished
ror.fileinfo12261509170001831 13261509160001831 DATM_ A Novel Data Agnostic Topic Modeling Technique With Improved Effectiveness for Both Short and Long Text
ror.mmsid9916727825601831

Files

Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
DATM_ A Novel Data Agnostic Topic Modeling Technique With Improved Effectiveness for Both Short and Long Text.pdf
Size:
1.63 MB
Format:
Adobe Portable Document Format
Description:
Published version
Download

Collections