基于贝叶斯粗糙集的文本特征选择方法

特征选择是文本分类的一个核心研究课题.首先给出了一个基于最小词频的文档频,然后简单分析了经典粗糙集和变精度粗糙集的不足,紧接着把贝叶斯粗糙集引入进来并提出了一个属性约简算法,最后把该属性约简算法同基于最小词频的文档频结合起来,提出了一个综合的特征选择方法.该综合方法首先利用基于最小词频的文档频提取初始特征,然后利用所提属性约简算法消除冗余,从而获得较具代表性的特征子集.实验结果表明,该算法是有效的.

Text Feature Selection Method Based on Bayesian Rough Sets

Feature selection is the core research topic in text categorization.It firstly presents document frequency based on minimum word frequency,and then simply analyzes shortcomings of classical rough sets and variable precision rough sets.It also introduces bayesian rough sets and presents an algorithm of attribute reduction.Finally,combined the attribute reduction algorithm with document frequency based on minimum word frequency it proposes a comprehensive feature selection method.The comprehensive method firstly uses document frequency method based on minimum word frequency to extract original features,and then employs the attribute reduction algorithm to eliminate redundancy.Experimental results show that the algorithm is effective.