This method USES kernel density estimation model to construct the approximate density function, and takes hill climbing strategy to extract clustering patterns.
该方法采用核密度估计模型来构造近似密度函数，利用爬山策略来提取聚类模式。
2
The main idea is to approximate the classical local linear embedding (LLE) by introducing a linear transformation matrix and then find the solution in a very high dimensional space by kernel trick.
According to the characteristic of SVDD, the proposed algorithm utilizes the non-Gaussian to measure how kernel samples approximate to a spherical area, and then optimize the kernel parameter.