{"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/learning-deep-structured-semantic-models-for","title":"Learning deep structured semantic models for web search using clickthrough data","arxiv_id":null,"date":"2013-10-27","proceeding":"CIKM 2013 10","authors":["Po-Sen Huang","Xiaodong He","Jianfeng Gao","Li Deng","Alex Acero","Larry Heck"],"abstract":"Latent semantic models, such as LSA, intend to map a query to its\r\nrelevant documents at the semantic level where keyword-based\r\nmatching often fails. In this study we strive to develop a series of\r\nnew latent semantic models with a deep structure that project\r\nqueries and documents into a common low-dimensional space\r\nwhere the relevance of a document given a query is readily\r\ncomputed as the distance between them. The proposed deep\r\nstructured semantic models are discriminatively trained by\r\nmaximizing the conditional likelihood of the clicked documents\r\ngiven a query using the clickthrough data. To make our models\r\napplicable to large-scale Web search applications, we also use a\r\ntechnique called word hashing, which is shown to effectively\r\nscale up our semantic models to handle large vocabularies which\r\nare common in such tasks. The new models are evaluated on a\r\nWeb document ranking task using a real-world data set. Results\r\nshow that our best model significantly outperforms other latent\r\nsemantic models, which were considered state-of-the-art in the\r\nperformance prior to the work presented in this paper.","url_abs":"https://dl.acm.org/doi/10.1145/2505515.2505665","url_pdf":"https://posenhuang.github.io/papers/cikm2013_DSSM_fullversion.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":"learning-deep-structured-semantic-models-for","repo_url":"https://github.com/PaddlePaddle/PaddleRec/tree/release/2.1.0/models/match/dssm","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"paddle","reach":null},{"paper_slug":"learning-deep-structured-semantic-models-for","repo_url":"https://github.com/UlionTse/mlgb","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"learning-deep-structured-semantic-models-for","repo_url":"https://github.com/Wings236/DSSM_pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"learning-deep-structured-semantic-models-for","repo_url":"https://github.com/alibaba/EasyRec/blob/master/easy_rec/python/model/dssm.py","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":null},{"paper_slug":"learning-deep-structured-semantic-models-for","repo_url":"https://github.com/alibaba/TorchEasyRec","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"learning-deep-structured-semantic-models-for","repo_url":"https://github.com/xue-pai/FuxiCTR","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"document-ranking","task_name":"Document Ranking"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}