%0 Conference Paper %1 zhang04domain %A Zhang, Zhuo %A Sylva, Lyne Da %A Davidson, Colin %A Lizarralde, Gonzalo %A Nie, Jian-Yun %B Proceedings of the SIGIR 04 Workshop: Information Retrieval for Question Answering %C Sheffield, UK %D 2004 %K 2004 domain gate nlp NT2OD %T Domain-Specific QA for the Construction Sector %U http://nlp.shef.ac.uk/ir4qa04/ZhangEtAlFinal.pdf %X Previous research on Question-Answering (QA) has focused on general domain questions. The general approach is to first recognize Named Entities (NE) in both texts and questions; then the most relevant answers (or passages) are selected (after an IR selection) according to the type of question and the NEs included in the possible answers. In this paper, we extend this general approach to domain-specific questions in the construction sector. This extension allows us to answer questions such as “What material is appropriate for ...”, which cannot be answered by a general QA system. Our approach is based on a domain-specific thesaurus, which contains a large set of domain-specific concepts organized into a hierarchy. Generic concepts such as “material” are considered as semantic categories. Our experiments on a technical corpus in construction show that this approach is effective: using our extension, we can obtain improvements on Mean Reciprocal Rank (MRR) of about 10%.

@inproceedings{zhang04domain, abstract = {Previous research on Question-Answering (QA) has focused on general domain questions. The general approach is to first recognize Named Entities (NE) in both texts and questions; then the most relevant answers (or passages) are selected (after an IR selection) according to the type of question and the NEs included in the possible answers. In this paper, we extend this general approach to domain-specific questions in the construction sector. This extension allows us to answer questions such as “What material is appropriate for ...”, which cannot be answered by a general QA system. Our approach is based on a domain-specific thesaurus, which contains a large set of domain-specific concepts organized into a hierarchy. Generic concepts such as “material” are considered as semantic categories. Our experiments on a technical corpus in construction show that this approach is effective: using our extension, we can obtain improvements on Mean Reciprocal Rank (MRR) of about 10%.}, added-at = {2007-02-25T19:59:56.000+0100}, address = {Sheffield, UK}, author = {Zhang, Zhuo and Sylva, Lyne Da and Davidson, Colin and Lizarralde, Gonzalo and Nie, Jian-Yun}, biburl = {https://www.bibsonomy.org/bibtex/2d00009712ecedd21bf99ce90c567d0c3/butonic}, booktitle = {Proceedings of the SIGIR 04 Workshop: Information Retrieval for Question Answering}, day = 29, description = {SIGIR 04 Workshop: Information Retrieval for Question Answering}, interhash = {bc2396453318ac5c03add5e85a404b6b}, intrahash = {d00009712ecedd21bf99ce90c567d0c3}, keywords = {2004 domain gate nlp NT2OD}, month = Jul, timestamp = {2007-02-25T19:59:56.000+0100}, title = {{D}omain-{S}pecific {QA} for the {C}onstruction {S}ector }, url = {http://nlp.shef.ac.uk/ir4qa04/ZhangEtAlFinal.pdf}, year = 2004 }