CAJ | 학술논문

利用核函数主元分析(KPCA)方法对大样本、高维数据进行特征提取预处理,并结合文化算法(CA)选择最优或接近最优的核函数,将其用于模糊C均值(FCM)聚类中,不但有效地提取了样本的非线性信息,而且使样本维数得到约简.实验表明该方法具有较好的聚类效果和更少的训练时间.
이용핵함수주원분석(KPCA)방법대대양본、고유수거진행특정제취예처리,병결합문화산법(CA)선택최우혹접근최우적핵함수,장기용우모호C균치(FCM)취류중,불단유효지제취료양본적비선성신식,이차사양본유수득도약간.실험표명해방법구유교호적취류효과화경소적훈련시간.
Kernel PCA method extracts feature from large samples and high dimension data sets,combining CA to select optimized kernel function or near optimized kernel function.FCM based on the method not only effectively extracts the nonlinear information from the samples but also reduces dimension.Experiment shows its better clustering result and less train time.