{"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/nine-features-in-a-random-forest-to-learn","title":"Nine Features in a Random Forest to Learn Taxonomical Semantic Relations","arxiv_id":"1603.08702","date":"2016-03-29","proceeding":"LREC 2016 5","authors":["Enrico Santus","Alessandro Lenci","Tin-Shing Chiu","Qin Lu","Chu-Ren Huang"],"abstract":"ROOT9 is a supervised system for the classification of hypernyms, co-hyponyms\nand random words that is derived from the already introduced ROOT13 (Santus et\nal., 2016). It relies on a Random Forest algorithm and nine unsupervised\ncorpus-based features. We evaluate it with a 10-fold cross validation on 9,600\npairs, equally distributed among the three classes and involving several\nParts-Of-Speech (i.e. adjectives, nouns and verbs). When all the classes are\npresent, ROOT9 achieves an F1 score of 90.7%, against a baseline of 57.2%\n(vector cosine). When the classification is binary, ROOT9 achieves the\nfollowing results against the baseline: hypernyms-co-hyponyms 95.7% vs. 69.8%,\nhypernyms-random 91.8% vs. 64.1% and co-hyponyms-random 97.8% vs. 79.4%. In\norder to compare the performance with the state-of-the-art, we have also\nevaluated ROOT9 in subsets of the Weeds et al. (2014) datasets, proving that it\nis in fact competitive. Finally, we investigated whether the system learns the\nsemantic relation or it simply learns the prototypical hypernyms, as claimed by\nLevy et al. (2015). The second possibility seems to be the most likely, even\nthough ROOT9 can be trained on negative examples (i.e., switched hypernyms) to\ndrastically reduce this bias.","url_abs":"http://arxiv.org/abs/1603.08702v1","url_pdf":"http://arxiv.org/pdf/1603.08702v1.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":"nine-features-in-a-random-forest-to-learn","repo_url":"https://github.com/esantus/ROOT9","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"classification","task_name":"General Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1603.08702","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}