{"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/familia-a-configurable-topic-modeling","title":"Familia: A Configurable Topic Modeling Framework for Industrial Text Engineering","arxiv_id":"1808.03733","date":"2018-08-11","proceeding":null,"authors":["Di Jiang","Yuanfeng Song","Rongzhong Lian","Siqi Bao","Jinhua Peng","Huang He","Hua Wu"],"abstract":"In the last decade, a variety of topic models have been proposed for text\nengineering. However, except Probabilistic Latent Semantic Analysis (PLSA) and\nLatent Dirichlet Allocation (LDA), most of existing topic models are seldom\napplied or considered in industrial scenarios. This phenomenon is caused by the\nfact that there are very few convenient tools to support these topic models so\nfar. Intimidated by the demanding expertise and labor of designing and\nimplementing parameter inference algorithms, software engineers are prone to\nsimply resort to PLSA/LDA, without considering whether it is proper for their\nproblem at hand or not. In this paper, we propose a configurable topic modeling\nframework named Familia, in order to bridge the huge gap between academic\nresearch fruits and current industrial practice. Familia supports an important\nline of topic models that are widely applicable in text engineering scenarios.\nIn order to relieve burdens of software engineers without knowledge of Bayesian\nnetworks, Familia is able to conduct automatic parameter inference for a\nvariety of topic models. Simply through changing the data organization of\nFamilia, software engineers are able to easily explore a broad spectrum of\nexisting topic models or even design their own topic models, and find the one\nthat best suits the problem at hand. With its superior extendability, Familia\nhas a novel sampling mechanism that strikes balance between effectiveness and\nefficiency of parameter inference. Furthermore, Familia is essentially a big\ntopic modeling framework that supports parallel parameter inference and\ndistributed parameter storage. The utilities and necessity of Familia are\ndemonstrated in real-life industrial applications. Familia would significantly\nenlarge software engineers' arsenal of topic models and pave the way for\nutilizing highly customized topic models in real-life problems.","url_abs":"http://arxiv.org/abs/1808.03733v2","url_pdf":"http://arxiv.org/pdf/1808.03733v2.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":"familia-a-configurable-topic-modeling","repo_url":"https://github.com/baidu/Familia","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"paddle","reach":null}],"tasks":[{"task_slug":"topic-models","task_name":"Topic Models"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}