摘要
特征参数分类泛化性差及分类计算量大影响着K近邻(KNN)的分类性能。提出了一种降维条件下基于联合熵的改进KNN算法,其具体思路是,通过计算任意两个条件属性下对应的特征参数的联合熵衡量数据特征针对分类影响程度的大小,建立特征分类特性与具体分类过程的内在联系,并给出根据特征联合熵集约简条件属性的方法。理论分析与仿真实验表明,与经典KNN等算法相比,提出的算法具有更高的分类性能。
Poor generalization of feature parameters classification and large category computation reduce the classification performace of K-Nearest Neighbor（KNN）.An improved KNN based on union entropy under the attribute reduction condition was proposed.Firstly,the size of classification impact of data feature was measured by calculating the union entropy of two feature parameters relative to any two condition attributes,and the intrinsic relation was established between classified features and the specific classification process.Then,the method which reduced condition attributes according feature union entropy set was given.The theoretical analysis and the simulation experiment show that compared with the classical KNN,the improved algorithm has better classification performance.
出处
《计算机应用》
CSCD
北大核心
2011年第7期1785-1788,1792,共5页
journal of Computer Applications
关键词
K近邻
特征
联合熵
条件属性
分类
K-Nearest Neighbor（KNN）
feature
union entropy
condition attribute
classification
作者简介
作者简介：周靖（1980-），女，广东茂名人，实验师，硕士，主要研究方向：人工智能、数据挖掘；（zhou_jing1980@126．com）
刘晋胜（1979-），男，广东梅州人，实验师，硕士，主要研究方向：人工智能、嵌入式系统、信号系统处理。
