{"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/neural-vector-spaces-for-unsupervised","title":"Neural Vector Spaces for Unsupervised Information Retrieval","arxiv_id":"1708.02702","date":"2017-08-09","proceeding":null,"authors":["Christophe Van Gysel","Maarten de Rijke","Evangelos Kanoulas"],"abstract":"We propose the Neural Vector Space Model (NVSM), a method that learns\nrepresentations of documents in an unsupervised manner for news article\nretrieval. In the NVSM paradigm, we learn low-dimensional representations of\nwords and documents from scratch using gradient descent and rank documents\naccording to their similarity with query representations that are composed from\nword representations. We show that NVSM performs better at document ranking\nthan existing latent semantic vector space methods. The addition of NVSM to a\nmixture of lexical language models and a state-of-the-art baseline vector space\nmodel yields a statistically significant increase in retrieval effectiveness.\nConsequently, NVSM adds a complementary relevance signal. Next to semantic\nmatching, we find that NVSM performs well in cases where lexical matching is\nneeded.\n  NVSM learns a notion of term specificity directly from the document\ncollection without feature engineering. We also show that NVSM learns\nregularities related to Luhn significance. Finally, we give advice on how to\ndeploy NVSM in situations where model selection (e.g., cross-validation) is\ninfeasible. We find that an unsupervised ensemble of multiple models trained\nwith different hyperparameter values performs better than a single\ncross-validated model. Therefore, NVSM can safely be used for ranking documents\nwithout supervised relevance judgments.","url_abs":"http://arxiv.org/abs/1708.02702v4","url_pdf":"http://arxiv.org/pdf/1708.02702v4.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":"neural-vector-spaces-for-unsupervised","repo_url":"https://github.com/cvangysel/cuNVSM","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"neural-vector-spaces-for-unsupervised","repo_url":"https://github.com/osirrc/nvsm-docker","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"neural-vector-spaces-for-unsupervised","repo_url":"https://github.com/osirrc2019/nvsm-docker","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"neural-vector-spaces-for-unsupervised","repo_url":"https://github.com/rodgzilla/NVSM_pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"document-ranking","task_name":"Document Ranking"},{"task_slug":"feature-engineering","task_name":"Feature Engineering"},{"task_slug":"information-retrieval","task_name":"Information Retrieval"},{"task_slug":"model-selection","task_name":"Model Selection"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"specificity","task_name":"Specificity"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1708.02702","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}