Title :
Research on Topic Relevancy of Sentences Based on HowNet Semantic Computation
Author :
Xu, Jinzhong ; Liu, Jie ; Liu, Xiaoming
Author_Institution :
Beijing Inst. of Technol., BIT Beijing, Beijing Inst. of Technol., Beijing, China
Abstract :
In Automatic Question Answering System, topic identification is usually based on relevancy computation of sentences. This paper introduces an approach to compute topic relevancy. Using the semantic computation in HowNet, the relevancy of sentences can be calculated. The topic relevancy of sentences can be calculated through the computation of relevancy between words of sentence and subject words. Experimental results show the effectiveness of the method.
Keywords :
ontologies (artificial intelligence); text analysis; HowNet semantic computation; automatic question answering system; sentence relevancy computation; sentence topic relevancy; topic identification; Classification tree analysis; Computer science; Dictionaries; Educational institutions; Hybrid intelligent systems; Knowledge engineering; Natural language processing; Ontologies; Reflection; Software libraries; Automatic Question Answering System; Hownet; semantic computation; topic identification; topic relevancy;
Conference_Titel :
Hybrid Intelligent Systems, 2009. HIS '09. Ninth International Conference on
Conference_Location :
Shenyang
Print_ISBN :
978-0-7695-3745-0
DOI :
10.1109/HIS.2009.150
