一种基于投票机制的代表点选择算法  

A prototype selection algorithm based on voting mechanism

作　　者：任江涛[1] 丘正元[1] 纪庆革[1] 

机构地区：[1]中山大学计算机科学系,广东广州510275

出　　处：《计算机应用》2007年第1期77-79,共3页journal of Computer Applications

基　　金：国家自然科学基金资助项目(60473109);广东省自然科学基金资助项目(0430046204300602)

摘　　要：代表点选择是面向数据挖掘与模式识别的数据预处理的重要内容之一,是提高分类器分类正确率和执行效率的重要途径。提出了一种基于投票机制的代表点选择算法,该算法能使所得到的代表点尽可能分布在类别边界上,且投票选择机制易于排除异常点,减少数据量,从而有利于提高最近邻分类器的分类精度和效率。通过与多个经典的代表点选择算法的实验比较分析,表明所提出的基于投票机制的代表点选择算法在提高最近邻分类器分类精度和数据降低率上都具有一定的优势。Prototype selection is an important step in data mining and pattern recognition; it is an efficient way to improve the classification accuracy and runing efficiency. This paper proposed a new prototype selection algorithm based on voting mechanism, which could see that most of the selected prototypes are distributed among the classification border and are not outliers. And it is in favor of improving the classification accuracy and efficiency of the nearest neighbor classifier. Experiments on the proposed algorithm and some other famous prototype selection algorithms show the prototype selection algorithm based on voting mechanism has some advantages in improving both the classification accuracy and the data reduction rate.

关 键 词：代表点选择 最近邻分类 投票机制 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论] TP18[自动化与计算机技术—计算机科学与技术]