{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/stability-of-topic-modeling-via-matrix","title":"Stability of Topic Modeling via Matrix Factorization","arxiv_id":"1702.07186","date":"2017-02-23","proceeding":null,"authors":["Mark Belford","Brian Mac Namee","Derek Greene"],"abstract":"Topic models can provide us with an insight into the underlying latent\nstructure of a large corpus of documents. A range of methods have been proposed\nin the literature, including probabilistic topic models and techniques based on\nmatrix factorization. However, in both cases, standard implementations rely on\nstochastic elements in their initialization phase, which can potentially lead\nto different results being generated on the same corpus when using the same\nparameter values. This corresponds to the concept of \"instability\" which has\npreviously been studied in the context of $k$-means clustering. In many\napplications of topic modeling, this problem of instability is not considered\nand topic models are treated as being definitive, even though the results may\nchange considerably if the initialization process is altered. In this paper we\ndemonstrate the inherent instability of popular topic modeling approaches,\nusing a number of new measures to assess stability. To address this issue in\nthe context of matrix factorization for topic modeling, we propose the use of\nensemble learning strategies. Based on experiments performed on annotated text\ncorpora, we show that a K-Fold ensemble strategy, combining both ensembles and\nstructured initialization, can significantly reduce instability, while\nsimultaneously yielding more accurate topic models.","url_abs":"http://arxiv.org/abs/1702.07186v2","url_pdf":"http://arxiv.org/pdf/1702.07186v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"stability-of-topic-modeling-via-matrix","repo_url":"https://github.com/derekgreene/topic-ensemble","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"clustering","task_name":"Clustering"},{"task_slug":"ensemble-learning","task_name":"Ensemble Learning"},{"task_slug":"topic-models","task_name":"Topic Models"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}