Benchmarking Sentence Embeddings in Textual Stream Clustering with Applications to Campaign Detection

Stampe, Lucas; Lütke-Stockdiek, Janina; Grimme, Britta; Grimme, Christian

Research article in edited proceedings (conference) | Peer reviewed

Abstract

Motivated by the emergence of large language models, we conduct a benchmark of sentence embeddings used to represent short texts in textual stream clustering. We achieve comparable results by adapting a non-textual stream clustering algorithm to use sentence embeddings compared to textual stream clustering approaches that use other textual representation mechanisms. Benchmarking datasets with differing degrees of preprocessing are used. The results suggest that the chosen approach using sentence embeddings does not perform as well as previous approaches on preprocessed datasets but has more significant potential on less preprocessed datasets. This highlights the need for new and more application-oriented benchmarking datasets for stream clustering. Further, we conduct a case study in the context of social media campaign detection and show that the approaches are able to find traces of orchestrated activities.

Details about the publication

PublisherHirose, Akira; Ishibuchi, Hisao; Jayne, Chrisina;
Book titleProceedings of the IEEE World Congress on Computational Intelligence (WCCI) - International Joint Conference on Neural Networks (IJCNN)
Page range1-8
Publishing companyWiley-IEEE Press
Place of publicationNew Jersey
StatusPublished
Release year2024
Language in which the publication is writtenEnglish
ConferenceIEEE World Congress on Computational Intelligence, Yokohama, Japan
ISBN979-8-3503-5931-2
DOI10.1109/IJCNN60899.2024.10650595
Link to the full texthttps://ieeexplore.ieee.org/abstract/document/10650595
Keywordsstream clustering; embeddings; benchmark

Authors from the University of Münster

Grimme, Christian
Research Group Computational Social Science and Systems Analysis (CSSSA)
Lütke-Stockdiek, Janina Susanne
Research Group Computational Social Science and Systems Analysis (CSSSA)
Stampe, Lucas
Research Group Computational Social Science and Systems Analysis (CSSSA)