{"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/computing-web-scale-topic-models-using-an","title":"Computing Web-scale Topic Models using an Asynchronous Parameter Server","arxiv_id":"1605.07422","date":"2016-05-24","proceeding":null,"authors":["Rolf Jagerman","Carsten Eickhoff","Maarten de Rijke"],"abstract":"Topic models such as Latent Dirichlet Allocation (LDA) have been widely used\nin information retrieval for tasks ranging from smoothing and feedback methods\nto tools for exploratory search and discovery. However, classical methods for\ninferring topic models do not scale up to the massive size of today's publicly\navailable Web-scale data sets. The state-of-the-art approaches rely on custom\nstrategies, implementations and hardware to facilitate their asynchronous,\ncommunication-intensive workloads.\n  We present APS-LDA, which integrates state-of-the-art topic modeling with\ncluster computing frameworks such as Spark using a novel asynchronous parameter\nserver. Advantages of this integration include convenient usage of existing\ndata processing pipelines and eliminating the need for disk writes as data can\nbe kept in memory from start to finish. Our goal is not to outperform highly\ncustomized implementations, but to propose a general high-performance topic\nmodeling framework that can easily be used in today's data processing\npipelines. We compare APS-LDA to the existing Spark LDA implementations and\nshow that our system can, on a 480-core cluster, process up to 135 times more\ndata and 10 times more topics without sacrificing model quality.","url_abs":"http://arxiv.org/abs/1605.07422v3","url_pdf":"http://arxiv.org/pdf/1605.07422v3.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":"computing-web-scale-topic-models-using-an","repo_url":"https://github.com/rjagerman/glint","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"information-retrieval","task_name":"Information Retrieval"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"topic-models","task_name":"Topic Models"}],"methods":[{"method_slug":"lda","method_name":"LDA"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}