In-domain relation discovery with meta-constraints via posterior regularization
We present a novel approach to discovering relations and their instantiations from a collection of documents in a single domain. Our approach learns relation types by exploiting meta-constraints that characterize the general qualities of a good relation in any domain. These constraints state that in...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
Association for Computing Machinery
2012
|
| Online Access: | http://hdl.handle.net/1721.1/73079 https://orcid.org/0000-0002-2921-8201 |
| _version_ | 1826209117326278656 |
|---|---|
| author | Chen, Harr Benson, Edward Oscar Naseem, Tahira Barzilay, Regina |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Chen, Harr Benson, Edward Oscar Naseem, Tahira Barzilay, Regina |
| author_sort | Chen, Harr |
| collection | MIT |
| description | We present a novel approach to discovering relations and their instantiations from a collection of documents in a single domain. Our approach learns relation types by exploiting meta-constraints that characterize the general qualities of a good relation in any domain. These constraints state that instances of a single relation should exhibit regularities at multiple levels of linguistic structure, including lexicography, syntax, and document-level context. We capture these regularities via the structure of our probabilistic model as well as a set of declaratively-specified constraints enforced during posterior inference. Across two domains our approach successfully recovers hidden relation structure, comparable to or outperforming previous state-of-the-art approaches. Furthermore, we find that a small set of constraints is applicable across the domains, and that using domain-specific constraints can further improve performance. |
| first_indexed | 2024-09-23T14:17:51Z |
| format | Article |
| id | mit-1721.1/73079 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T14:17:51Z |
| publishDate | 2012 |
| publisher | Association for Computing Machinery |
| record_format | dspace |
| spelling | mit-1721.1/730792022-10-01T20:25:05Z In-domain relation discovery with meta-constraints via posterior regularization Chen, Harr Benson, Edward Oscar Naseem, Tahira Barzilay, Regina Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Barzilay, Regina Chen, Harr Barzilay, Regina Benson, Edward Oscar Naseem, Tahira We present a novel approach to discovering relations and their instantiations from a collection of documents in a single domain. Our approach learns relation types by exploiting meta-constraints that characterize the general qualities of a good relation in any domain. These constraints state that instances of a single relation should exhibit regularities at multiple levels of linguistic structure, including lexicography, syntax, and document-level context. We capture these regularities via the structure of our probabilistic model as well as a set of declaratively-specified constraints enforced during posterior inference. Across two domains our approach successfully recovers hidden relation structure, comparable to or outperforming previous state-of-the-art approaches. Furthermore, we find that a small set of constraints is applicable across the domains, and that using domain-specific constraints can further improve performance. United States. Defense Advanced Research Projects Agency (Machine Reading Program under Air Force Research Laboratory (AFRL) prime contract no. FA8750-09-C-0172) 2012-09-20T18:04:40Z 2012-09-20T18:04:40Z 2011-06 Article http://purl.org/eprint/type/ConferencePaper 978-1-932432-87-9 http://hdl.handle.net/1721.1/73079 Chen, Harr et al. "In-domain Relation Discovery with Meta-constraints via Posterior Regularization." Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies - Volume 1, HLT '11, Portland, Oregon, USA, June 19-24, 2011. https://orcid.org/0000-0002-2921-8201 en_US http://dl.acm.org/citation.cfm?id=2002472.2002540&coll=DL&dl=ACM&CFID=87070219&CFTOKEN=34670296 Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies - Volume 1, ACL HLT '11 Creative Commons Attribution-Noncommercial-Share Alike 3.0 http://creativecommons.org/licenses/by-nc-sa/3.0/ application/pdf Association for Computing Machinery MIT web domain |
| spellingShingle | Chen, Harr Benson, Edward Oscar Naseem, Tahira Barzilay, Regina In-domain relation discovery with meta-constraints via posterior regularization |
| title | In-domain relation discovery with meta-constraints via posterior regularization |
| title_full | In-domain relation discovery with meta-constraints via posterior regularization |
| title_fullStr | In-domain relation discovery with meta-constraints via posterior regularization |
| title_full_unstemmed | In-domain relation discovery with meta-constraints via posterior regularization |
| title_short | In-domain relation discovery with meta-constraints via posterior regularization |
| title_sort | in domain relation discovery with meta constraints via posterior regularization |
| url | http://hdl.handle.net/1721.1/73079 https://orcid.org/0000-0002-2921-8201 |
| work_keys_str_mv | AT chenharr indomainrelationdiscoverywithmetaconstraintsviaposteriorregularization AT bensonedwardoscar indomainrelationdiscoverywithmetaconstraintsviaposteriorregularization AT naseemtahira indomainrelationdiscoverywithmetaconstraintsviaposteriorregularization AT barzilayregina indomainrelationdiscoverywithmetaconstraintsviaposteriorregularization |