Title :
A Chinese QA system based on lexical chain
Author :
Wang, Zhen-yu ; Zheng, Ying-yan
Author_Institution :
Sch. of Comput. Sci. & Eng., South China Univ. of Technol., Guangzhou
Abstract :
Most technologies of Chinese question answering system did not make full use of the semantic information. This paper introduces an idea of building QA system based on lexical chain. The approach is to build lexical chains between question and the candidate answers, and then elect the most exact answer by comparing the chain connectivity. In the process, HowNet was used to analyze semantic roles of the key components in the question sentence. This method makes use of the semantic information in a deeper level, and improves the performance of QA system.
Keywords :
information retrieval; natural language processing; Chinese question answering system; HowNet; chain connectivity; lexical chain; question sentence; semantic information; Computer science; Cybernetics; Data mining; Databases; Information retrieval; Machine learning; Natural languages; Pattern matching; Search engines; Web pages; Chinese Question Answering System; HowNet; Lexical chain; Semantic role analysis;
Conference_Titel :
Machine Learning and Cybernetics, 2008 International Conference on
Conference_Location :
Kunming
Print_ISBN :
978-1-4244-2095-7
Electronic_ISBN :
978-1-4244-2096-4
DOI :
10.1109/ICMLC.2008.4620819
