摘要
问题分类是问答系统中重要的组成部分,问题分类结果的好坏直接影响问答系统的质量。本文提出了一种用于问题分类的特征提取的新方法,该方法主要使用句法分析的结果,提取问题的主干和疑问词及其附属成分作为分类的特征,此方法大幅度地减少了噪音,突出了问题分类的主要特征,利用贝叶斯分类器分类,有效地提高了问题分类的精度。实验结果证明了该方法的有效性,大类和小类的分类精度分别达到了86.62%和71.92%,取得了较好的效果。
Qnestion classification is very important for question answering, and the result of question classification directly affects the quality of question answering. This paper presents a new method on feature extraction for question classification. The output of syntactic parsing is used in this method to extract the Subject-Predieate structure as well as interrogative words and their adjunctive parts as features for elassifieation, leading to substantial roduetion in noise, and emphasis on the main features of question claasification. A bayesian claasifier is used in classification, which effectively increases the precision of question claasifieation. The experimental result validates the effeetiveness of this method： the classifieation precision of coarse classes and fine elasses reach 86.62% and 71.92% respectively, which attains the expected effects.
出处
《中文信息学报》
CSCD
北大核心
2006年第2期33-39,共7页
Journal of Chinese Information Processing
基金
国家自然科学基金资助项目(60435020)
关键词
计算机应用
中文信息处理
问答系统
问题分类
特征提取
句法分析
computer application
Chinese information processing
question answering system
question classification
feature extraction
syntactic parsing
