Ontology-Based Semantic Q&A system in Health Care: An IllustratedApplication on Down Syndrome

• Main Authors: Shu-Hai Wang, 王書海
• Other Authors: Keng-Pei Lin
• Format: Others
• Language: en_US
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/2n9td4

碩士 === 國立中山大學 === 資訊管理學系研究所 === 105 === This study presents an ontology-based Semantic Question and Answer (Q&A) system applied to the Down Syndrome in Healthcare. We proposed a Q&A system which allows for users to ask questions in natural language, and the system will search the answers from the ontology by reasoning with related keywords from the natural language searched. In order to figure out what users’ questions meant we use the Classified Knowledge and Information Processing (CKIP) to tokenize words and to tag Part-Of-Speech (POS) in the questions. This Q&A system also uses a combination of three keywords, including medical terminologies, intention words which only exist in the domain of Medicine, and words of 5W1H, to analyze natural language questions. These three kinds of keywords allow our approach of answer extraction to reason in the ontology by formulated SPARQL queries. This system has been developed and tested in the Chinese language. The ontology is being adopted for classifying Down syndrome related information. We have carried out experiments to evaluate our approach of question analysis and answer extraction. The preliminary result shows that the excellent performance of our proposed approach.