{"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/anchored-correlation-explanation-topic","title":"Anchored Correlation Explanation: Topic Modeling with Minimal Domain Knowledge","arxiv_id":"1611.10277","date":"2016-11-30","proceeding":"TACL 2017 1","authors":["Ryan J. Gallagher","Kyle Reing","David Kale","Greg Ver Steeg"],"abstract":"While generative models such as Latent Dirichlet Allocation (LDA) have proven\nfruitful in topic modeling, they often require detailed assumptions and careful\nspecification of hyperparameters. Such model complexity issues only compound\nwhen trying to generalize generative models to incorporate human input. We\nintroduce Correlation Explanation (CorEx), an alternative approach to topic\nmodeling that does not assume an underlying generative model, and instead\nlearns maximally informative topics through an information-theoretic framework.\nThis framework naturally generalizes to hierarchical and semi-supervised\nextensions with no additional modeling assumptions. In particular, word-level\ndomain knowledge can be flexibly incorporated within CorEx through anchor\nwords, allowing topic separability and representation to be promoted with\nminimal human intervention. Across a variety of datasets, metrics, and\nexperiments, we demonstrate that CorEx produces topics that are comparable in\nquality to those produced by unsupervised and semi-supervised variants of LDA.","url_abs":"http://arxiv.org/abs/1611.10277v4","url_pdf":"http://arxiv.org/pdf/1611.10277v4.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":"anchored-correlation-explanation-topic","repo_url":"https://github.com/gregversteeg/corex_topic","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[],"methods":[{"method_slug":"lda","method_name":"LDA"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1611.10277","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}