{"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/word-representations-tree-models-and","title":"Word Representations, Tree Models and Syntactic Functions","arxiv_id":"1508.07709","date":"2015-08-31","proceeding":null,"authors":["Simon Šuster","Gertjan van Noord","Ivan Titov"],"abstract":"Word representations induced from models with discrete latent variables\n(e.g.\\ HMMs) have been shown to be beneficial in many NLP applications. In this\nwork, we exploit labeled syntactic dependency trees and formalize the induction\nproblem as unsupervised learning of tree-structured hidden Markov models.\nSyntactic functions are used as additional observed variables in the model,\ninfluencing both transition and emission components. Such syntactic information\ncan potentially lead to capturing more fine-grain and functional distinctions\nbetween words, which, in turn, may be desirable in many NLP applications. We\nevaluate the word representations on two tasks -- named entity recognition and\nsemantic frame identification. We observe improvements from exploiting\nsyntactic function information in both cases, and the results rivaling those of\nstate-of-the-art representation learning methods. Additionally, we revisit the\nrelationship between sequential and unlabeled-tree models and find that the\nadvantage of the latter is not self-evident.","url_abs":"http://arxiv.org/abs/1508.07709v2","url_pdf":"http://arxiv.org/pdf/1508.07709v2.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":"word-representations-tree-models-and","repo_url":"https://github.com/rug-compling/hmm-reps","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"named-entity-recognition-1","task_name":"Named Entity Recognition"},{"task_slug":"named-entity-recognition-ner","task_name":"Named Entity Recognition (NER)"},{"task_slug":"representation-learning","task_name":"Representation Learning"},{"task_slug":"named-entity-recognition","task_name":"named-entity-recognition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}