{"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/hypernyms-under-siege-linguistically","title":"Hypernyms under Siege: Linguistically-motivated Artillery for Hypernymy Detection","arxiv_id":"1612.04460","date":"2016-12-14","proceeding":"EACL 2017 4","authors":["Vered Shwartz","Enrico Santus","Dominik Schlechtweg"],"abstract":"The fundamental role of hypernymy in NLP has motivated the development of\nmany methods for the automatic identification of this relation, most of which\nrely on word distribution. We investigate an extensive number of such\nunsupervised measures, using several distributional semantic models that differ\nby context type and feature weighting. We analyze the performance of the\ndifferent methods based on their linguistic motivation. Comparison to the\nstate-of-the-art supervised methods shows that while supervised methods\ngenerally outperform the unsupervised ones, the former are sensitive to the\ndistribution of training instances, hurting their reliability. Being based on\ngeneral linguistic hypotheses and independent from training data, unsupervised\nmeasures are more robust, and therefore are still useful artillery for\nhypernymy detection.","url_abs":"http://arxiv.org/abs/1612.04460v2","url_pdf":"http://arxiv.org/pdf/1612.04460v2.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":"hypernyms-under-siege-linguistically","repo_url":"https://github.com/vered1986/UnsupervisedHypernymy","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"hypernym-discovery","task_name":"Hypernym Discovery"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/hypernym-discovery-on-general","task":"Hypernym Discovery","dataset":"General","model":"balAPInc","rank_in_archive_order":8,"of":8,"metrics":{"MAP":"1.36","MRR":"3.18","P@5":"1.30"},"uses_additional_data":false},{"leaderboard":"/sota/hypernym-discovery-on-medical-domain","task":"Hypernym Discovery","dataset":"Medical domain","model":"balAPInc","rank_in_archive_order":8,"of":8,"metrics":{"MAP":"0.91","MRR":"2.10","P@5":"1.08"},"uses_additional_data":false},{"leaderboard":"/sota/hypernym-discovery-on-music-domain","task":"Hypernym Discovery","dataset":"Music domain","model":"balAPInc","rank_in_archive_order":7,"of":7,"metrics":{"MAP":"1.95","MRR":"5.01","P@5":"2.15"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1612.04460","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}