Detail Performance analysis of combined methods of genetic algorithm and K-means clustering in determining the value of centroid (article of Journal of Physics: Conference Series, Volume 930, conference 1) Bibliografi Author: Putra, Adya Zizwan ; Zarlis, Muhammad ; Nababan, Erna Budhiarti Topik: Centroid; K-Means; GenClust; Chromosome Bahasa: (EN )     Penerbit: IOP Publishing     Tempat Terbit: Bristol    Tahun Terbit: 2017     Jenis: Article - diterbitkan di jurnal ilmiah internasional Fulltext: Adya_Zizwan_2017_J._Phys.%3A_Conf._Ser._930_012008.pdf (212.58KB; 0 download) [Informasi yang berkaitan dengan koleksi ini di internet] Abstract The determination of Centroid on K-Means Algorithm directly affects the quality of the clustering results. Determination of centroid by using random numbers has many weaknesses. The GenClust algorithm that combines the use of Genetic Algorithms and KMeans uses a genetic algorithm to determine the centroid of each cluster. The use of the GenClust algorithm uses 50% chromosomes obtained through deterministic calculations and 50% is obtained from the generation of random numbers. This study will modify the use of the GenClust algorithm in which the chromosomes used are 100% obtained through deterministic calculations. The results of this study resulted in performance comparisons expressed in Mean Square Error influenced by centroid determination on K-Means method by using GenClust method, modified GenClust method and also classic K-Means. Opini Anda Klik untuk menuliskan opini Anda tentang koleksi ini! Lihat Sejarah Pengadaan  Konversi Metadata   Kembali

Copyright © 2006, 2007 Unika Atma Jaya, all rights reserved