{"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/linear-ensembles-of-word-embedding-models","title":"Linear Ensembles of Word Embedding Models","arxiv_id":"1704.01419","date":"2017-04-05","proceeding":"WS 2017 5","authors":["Avo Muromägi","Kairit Sirts","Sven Laur"],"abstract":"This paper explores linear methods for combining several word embedding\nmodels into an ensemble. We construct the combined models using an iterative\nmethod based on either ordinary least squares regression or the solution to the\northogonal Procrustes problem.\n  We evaluate the proposed approaches on Estonian---a morphologically complex\nlanguage, for which the available corpora for training word embeddings are\nrelatively small. We compare both combined models with each other and with the\ninput word embedding models using synonym and analogy tests. The results show\nthat while using the ordinary least squares regression performs poorly in our\nexperiments, using orthogonal Procrustes to combine several word embedding\nmodels into an ensemble model leads to 7-10% relative improvements over the\nmean result of the initial models in synonym tests and 19-47% in analogy tests.","url_abs":"http://arxiv.org/abs/1704.01419v1","url_pdf":"http://arxiv.org/pdf/1704.01419v1.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":"linear-ensembles-of-word-embedding-models","repo_url":"https://github.com/Shujian2015/meta-embedding-paper-list","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"word-embeddings","task_name":"Word Embeddings"},{"task_slug":"regression-1","task_name":"regression"}],"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}