支持向量机与哈夫曼树实现多分类的研究

基于支持向量机和决策树的多分类方法存在错误累积问题,累积的错误往往使分类准确率下降,分类效果变差.在仔细分析了其产生错误累积原因的基础上,提出了基于哈夫曼树的支持向量机多分类方法.该方法首先将一个多分类问题分解为多个二分类问题,针对每个二分类问题使用支持向量机二分类方法解决;然后根据相异度来决策分类的优先顺序,构建基于哈夫曼树的支持向量机多分类模型;最后使用勒卡斯开源数据集进行验证,并将它与传统的支持向量机多分类方法进行实验比较.实验结果表明,新的方法在分类速度和分类精度上较传统的支持向量机多分类方法优越.

Research on Multi-class Classification Based on SVM and Huffman Tree

There exists error accumulation in the multi-classification method,based on support vector machines and decision trees.It tends to decrease classification accuracy and results in a bad classification.With a careful analysis of error accumulation,it proposes a new multi-classification method,based on Huffman Tree and SVM.It divided a multi-classification problem into multiple binary classification problems,and gave classification priority,depending on the dissimilarity.At last,through an experiment with Lecast open source data sets,it verified the effectiveness.The experimental results show that the new method is superior to the traditional multi-classification method in classification speed and classification accuracy.