中文信息学报基于改进贝叶斯模型的问题分类①.pdf

中　文 　信 　息 　学 　报 第 19 卷 第 2 期 　 JOURNAL OF CHINESE INFORMATION PROCESSING Vol 19 No2 ( ) 文章编号 :1003 - 0077 2005 02 - 0100 - 06 基于改进贝叶斯模型的问题分类① 张 　宇 ,刘 　挺 ,文 　勖 ( 哈尔滨工业大学 计算机学院信息检索研究室 ,黑龙江 哈尔滨 　150001) 摘要 :随着计算机及互联网络技术的发展 ,开放域问答系统越来越受到人们的关注 ,因为它能够给用户提 供相对简洁 、准确的结果 。开放域问答系统通常包括问题分类 、问题扩展 、搜索引擎 、答案抽取和答案选择五 个主要部分 。问题分类在问答系统中起着很重要的作用 ,它的准确性直接影响到最终抽取的答案的准确性 。 本文在对已有的贝叶斯分类方法进行分析的基础上 ,对该方法进行了改进 。为了验证该方法的效果 ,构造了 问题的训练集和测试集 。从实验结果可以看出 ,该方法在实际应用中获得了较好的效果 。 关键词 :计算机应用 ; 中文信息处理 ;贝叶斯模型 ; 问题分类 ; 问答系统 中图分类号 :TP391. 2 　　文献标识码 :A Modif ied Bayesian Model Based Question Classif ication ZHAN G Yu ,LIU Ting ,WEN Xu ( Information Retrieval laboratory ,School of computer science and technology ,Harbin institute of technology , Harbin ,Heilongj iang 150001 , China) Abstract :With the rapid development of Internet ,the Opendomain Question Answering system becomes more and more at tractive because the compact and exact result can be given by it . The Opendomain Question Answering system is composed of five parts ,that is ,question classification ,question expansion , search engine ,answer extraction and answer selection . The question classification plays an important role in the question answering system ,and affects the correctness of the question an swering system directly. In this paper ,a modified Bayesian model was introduced based on analysis of the Bayesian model . The training set and testing set were constructed to verify the effect of this model . Experiments showed that this method could achieve better results in practice . key words : computer application ; Chinese information processing ; bayesian model ; question classificati