摘要
本文借鉴支持向量机非线性分类思想,提出基于序加权算子的分类模型。首先,在序加权平均算子的基础上,给出序加权算子的概念,并提出空间两点基于序加权算子的距离;对序加权算子的一些性质进行了研究。其次,通过对序加权算子几何意义的刻画,结合支持向量机最大分类间隔思想,建立基于序加权算子的数据分类模型及算法。最后,通过实例验证所给模型及算法的有效性,报告了与其他算法进行对比的实验结果。
The classification model based on ordered weighted operator is proposed by using the nonlinear classification idea of support vector machine. Firstly, ordered weighted operator and ordered weighted distance are given based on order weighted averaging operator. And some properties of ordered weighted operator are studied. Secondly, combining with the idea of the optimal classification boundary of support vector machine and the geometric description of ordered weighted operator, the data classification model and algorithm based on ordered weighted operator are built. Finally, some given examples show the efficiency of the models for solving classification problems. The experimental results are reported after comparing with other algorithms.
出处
《数码设计》
2017年第3期20-26,共7页
Peak Data Science
基金
国家自然科学基金(No.11501435)
西安市科技局计划项目(Program No.CXY1441(2))
关键词
序加权算子
支持向量机
CHOQUET积分
粒子群
ordered weighted operator
support vector machine
choquet integral
particle swarm