摘要
二叉树支持向量机是解决多类分类问题的算法之一,在目前多类分类算法中总体性能较优,但仍存在分类速度及分类精度不高等缺点。针对这些缺点,论文对二叉树支持向量机算法在二叉树结构及分类顺序两个方面进行改进,提出了基于类间相似度量数的二叉树构造算法。实验结果表明,论文算法具有更高的分类速度和准确度,能更好地解决多类分类问题。
Binary tree support vector machine is one of the algorithms to solve multi-class classification problems.In the cur⁃rent multi-class classification algorithm,the overall performance is better,but there are still some shortcomings such as classifica⁃tion speed and classification accuracy.Aiming at these shortcomings,this paper improves the binary tree support vector machine al⁃gorithm in two aspects of binary tree structure and classification order,and proposes a binary tree construction algorithm based on similar metrics between classes.The experimental results show that the proposed algorithm has higher classification speed and accu⁃racy,and can better solve multi-class classification problems.
作者
宋晓婉
黄树成
SONG Xiaowan;HUANG Shucheng(School of Computer Science,Jiangsu University of Science and Technology,Zhenjiang 212003)
出处
《计算机与数字工程》
2020年第8期1835-1839,共5页
Computer & Digital Engineering
基金
国家自然科学基金项目(编号:61772244)资助。
关键词
二叉树
多类文本
支持向量机
类间相似度量数
binary tree
multi-class text
support vector machine
inter-class similar measure number