ACADSTAFF UGM

CREATION
Title : EMPIRICAL STUDY ON BANDWIDTH OPTIMIZATION FOR KERNEL PCA IN THE K-MEANS CLUSTERING OF NON-LINEARLY SEPARABLE DATA
Author :

CORI PITOY (1) Prof. Drs. Subanar, Ph.D. (2) Dr. Abdurakhman (3)
Date : 31 2017
Keyword : K-Means Clustering,KPCA Bandwidth,Validity,Entropy, Non-Linear Separable Dataset K-Means Clustering,KPCA Bandwidth,Validity,Entropy, Non-Linear Separable Dataset
Abstract : K-Means is a method of non-hierarchical data clustering, which partitions observations into k clusters so that observations with the same characteristics are grouped into the same cluster, while observations with different characteristics are grouped into other cluster. The advantages of this method are easy to apply, simple, and efficient, and its success has been proven empirically. The problem is when the data is nonlinearly separable. Overcoming the problem of non-linearly separable data can be done through a data extraction and dimension reduction using Kernel Principal Component Analysis (KPCA). The results of KPCA transformation were affected by the kernel type and the size of bandwidth parameters (), as a smoothing parameter. Calculation of K-Means clustering of Iris dataset, using 2 Principal Component (PC), Euclid distance and Gaussian kernel showed that the external validity (entropy) and internal validity (Sum Square Within) are better than the result of standard K-Means algorithm.
Group of Knowledge : Statistik
Original Language : English
Level : Internasional
Status :
Published
Document
No Title Document Type Action
1 Paper EMPIRICAL STUDY ON BANDWIDTH OPTIMIZATION.pdf
Document Type : [PAK] Full Dokumen
[PAK] Full Dokumen View
2 Similarity EMPIRICAL STUDY ON BANDWIDTH OPTIMIZATION FOR KERNEL PCA IN THE K-MEANS CLUSTERING OF NON-LINEARLY SEPARABLE DATA.pdf
Document Type : [PAK] Cek Similarity
[PAK] Cek Similarity View
3 Journal of Theoretical and Applied Information Technology - Oct Daftar Isi.pdf
Document Type : [PAK] Full Dokumen
[PAK] Full Dokumen View
4 CS-EMPIRICAL STUDY ON BANDWIDTH OPTIMIZATION FOR KERNEL PCA IN THE K-MEANS CLUSTERING OF NON-LINEARLY SEPARABLE DATA.pdf
Document Type : [PAK] Cek Similarity
[PAK] Cek Similarity View