Title :
Question matching based on Fuzzy set
Author :
Gu, Tianjiao ; Ren, Fuji
Author_Institution :
Fac. of Eng., Univ. of Tokushima, Tokushima
Abstract :
Sentence similarity computing plays an important role in the question answering (QA) system. Because there are many question expressions for one meaning, we present a new approach to match question based on fuzzy set. In this paper, we establish a library of standard questions. Each standard question is relative with a series of Keywords. The main focus of this paper lies with matching of standard questions and questions asked by users. An experimental system based on the proposed method has been built, and the results of our experiments shows the proposed method is effective for question matching.
Keywords :
fuzzy set theory; information retrieval; natural language processing; pattern matching; fuzzy set; question answering system; question matching; sentence similarity computing; Algorithm design and analysis; Fuzzy sets; Fuzzy systems; Libraries; Measurement standards; Natural language processing; Natural languages; Search engines; Telecommunication computing; Telecommunication standards; Question Answering system; question matching; sentence similarity;
Conference_Titel :
Natural Language Processing and Knowledge Engineering, 2008. NLP-KE '08. International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4244-4515-8
Electronic_ISBN :
978-1-4244-2780-2
DOI :
10.1109/NLPKE.2008.4906799
