基于线性规划的支持向量聚类算法

为了克服k-均值聚类算法容易受到数据空间分布影响的缺点,将线性规划下的一类支持向量机算法与K-均值聚类方法相结合提出一种支持向量聚类算法,该算法的每次循环都采用线性规划下的一类支持向量机进行运算.该算法实现简单,与二次规划下的支持向量机聚类算法相比,该算法能够大大减小计算的复杂性,而且能保持良好的聚类效果.与K-均值聚类算法、自组织映射聚类算法等进行仿真比较,人工数据和实际数据表明了该算法的有效性和可行性.

Support vector clustering algorithm based on linear programming

In order to prevent the distribution of data from affecting the K-means clustering algorithm, a support vector clustering algorithm is proposed based on kernel method, which is inspired by the classical K-means algorithm in which each cluster is iteratively refined using a one-class support vector machine based on linear programming.This method can be easily implemented, the computational complexity is greatly reduced, and good clustering effect is obtained.Compared with popular clustering algorithms, like ...