{"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/improving-discourse-relation-projection-to","title":"Improving Discourse Relation Projection to Build Discourse Annotated Corpora","arxiv_id":"1707.06357","date":"2017-07-20","proceeding":"RANLP 2017 9","authors":["Majid Laali","Leila Kosseim"],"abstract":"The naive approach to annotation projection is not effective to project\ndiscourse annotations from one language to another because implicit discourse\nrelations are often changed to explicit ones and vice-versa in the translation.\nIn this paper, we propose a novel approach based on the intersection between\nstatistical word-alignment models to identify unsupported discourse\nannotations. This approach identified 65% of the unsupported annotations in the\nEnglish-French parallel sentences from Europarl. By filtering out these\nunsupported annotations, we induced the first PDTB-style discourse annotated\ncorpus for French from Europarl. We then used this corpus to train a classifier\nto identify the discourse-usage of French discourse connectives and show a 15%\nimprovement of F1-score compared to the classifier trained on the non-filtered\nannotations.","url_abs":"http://arxiv.org/abs/1707.06357v1","url_pdf":"http://arxiv.org/pdf/1707.06357v1.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":"improving-discourse-relation-projection-to","repo_url":"https://github.com/mjlaali/Europarl-ConcoDisco","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":null,"task_name":"Relation"},{"task_slug":"translation","task_name":"Translation"},{"task_slug":"word-alignment","task_name":"Word Alignment"}],"methods":[],"datasets_introduced":[{"slug":"europarl-concodisco-dataset","name":"Europarl ConcoDisco Dataset","full_name":null}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}