Title :
A Method for Distinguishing Japanese Question Sentences Based on the Number of Queries
Author_Institution :
Dept. of Comput. Sci. & Syst. Anal., Nihon Univ., Tokyo, Japan
fDate :
March 31 2009-April 2 2009
Abstract :
Figuring out the number of queries contained in a Japanese question sentence is necessary and an important process in question answering systems. Proposed here is a heuristics- and machine learning-based method for distinguishing between single-query question and multiple-query question, and estimating the number of queries in a Japanese question sentence. Experiments show that our method works well.
Keywords :
information retrieval systems; learning (artificial intelligence); natural language processing; query processing; Japanese question sentence; heuristics-based method; machine learning-based method; multiple-query question; question answering system; single-query question; Computer science; Data mining; Educational institutions; Informatics; Information analysis; Learning systems; Search engines; Support vector machine classification; Support vector machines; Web search; Question Answering (QA); Support Vector Machine (SVM); parallel structure;
Conference_Titel :
Computer Science and Information Engineering, 2009 WRI World Congress on
Conference_Location :
Los Angeles, CA
Print_ISBN :
978-0-7695-3507-4
DOI :
10.1109/CSIE.2009.751
