{"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/controlled-experiments-for-word-embeddings","title":"Controlled Experiments for Word Embeddings","arxiv_id":"1510.02675","date":"2015-10-09","proceeding":null,"authors":["Benjamin J. Wilson","Adriaan M. J. Schakel"],"abstract":"An experimental approach to studying the properties of word embeddings is\nproposed. Controlled experiments, achieved through modifications of the\ntraining corpus, permit the demonstration of direct relations between word\nproperties and word vector direction and length. The approach is demonstrated\nusing the word2vec CBOW model with experiments that independently vary word\nfrequency and word co-occurrence noise. The experiments reveal that word vector\nlength depends more or less linearly on both word frequency and the level of\nnoise in the co-occurrence distribution of the word. The coefficients of\nlinearity depend upon the word. The special point in feature space, defined by\nthe (artificial) word with pure noise in its co-occurrence distribution, is\nfound to be small but non-zero.","url_abs":"http://arxiv.org/abs/1510.02675v2","url_pdf":"http://arxiv.org/pdf/1510.02675v2.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":"controlled-experiments-for-word-embeddings","repo_url":"https://github.com/ElMehdiBouamama/MBTI-Tweetouilles","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"word-embeddings","task_name":"Word Embeddings"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1510.02675","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}