文章基本信息

标题：PERSIAN QUESTION CLASSIFICATION USING HEADWORD AND SEMANTIC FEATURES
本地全文：下载
作者：AMIR ROUSTAEI ; HAMID RASTEGARI
期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2018
卷号：96
期号：21
出版社：Journal of Theoretical and Applied
摘要：Question classification is an important component in question answering systems. The task of question classifier is to assign a label, depending on the classification strategy, to written question in natural language. Features are essential elements to obtaining an accurate question classifier. Low accuracy at the fine-grained level is the main problem among classifiers. In this paper, in order to improve the accuracy of question classification, two new features such as question�s headword and related semantic words are introduced. If headword is correctly identified, then the accuracy of answer classification increases. On the other hand, semantic meaning of related words effects on accuracy of the answer classification for both coarse and fine grained classes. The result shows the contribution of the presented features in coarse- and fine-grained classification accuracy.
关键词：Question Answering; Questions Classification; Machine Learning; Feature Extraction; Headword; Coarse and Fine-Grained Classification