EMAIL: PASSWORD:
Front Office
UPT. PERPUSTAKAAN
Institut Teknologi Sepuluh Nopember Surabaya


Kampus ITS Sukolilo - Surabaya 60111

Phone : 031-5921733 , 5923623
Fax : 031-5937774
E-mail : libits@its.ac.id
Website : http://library.its.ac.id

Support (Customer Service) :
timit_perpus@its.ac.id




Welcome..guys!

Have a problem with your access?
Please, contact our technical support below:
LIVE SUPPORT


Davi Wahyuni


Tondo Indra Nyata


Anis Wulandari


Ansi Aflacha




ITS » Master Theses » Rekayasa Kualitas S2
Posted by dee@its.ac.id at 16/06/2011 10:39:23  •  2768 Views


KLASTERING VARIETAS PADI MENGGUNAKAN MODIFIKASI METODE K-MEANS BERBASIS ORDERED WEIGHTED AVERAGING OWA

RICE VARIETIES CLUSTERING USING K-MEANS CLUSTERING BASED ORDERED WEIGHTED AVERAGING OWA

Author :
ULYA, MILLATUL  ( 2508201005 )




ABSTRAK

Metode k-means berbasis Ordered Weighted Averaging OWA telah dikembangkan oleh Cheng dkk 2009 untuk menyelesaikan kasus klasifikasi dengan cara mengintegrasikan k-means dan OWA. K-means sebenarnya merupakan suatu metode klastering dan OWA adalah operator agregasi. OWA dapat mengurangi kompleksitas data eksperimental yang akan diklasterkan dan dapat mempertimbangkan keterkaitan antar kriteria dari data yang dianalisis. Dilihat dari sisi fungsi k-means dan persamaan OWA yang digunakan maka diduga k-means berbasis OWA Cheng dkk 2009 dapat digunakan untuk menyelesaikan kasus klastering dengan memodifikasi beberapa tahapannya. Dalam penelitian ini akan dilakukan modifikasi k-means berbasis OWA Cheng dkk 2009 dan mengaplikasikannya pada kasus klastering varietas padi. Penelitian ini bertujuan untuk 1 Menginventaris variabel yang berkaitan dengan preferensi konsumen dalam memilih beras 2 Mengaplikasikan k-means berbasis OWA pada klastering data set iris untuk validasi dan pada data set padi 3 Mengukur tingkat akurasi metode k-means OWA dalam klastering data iris 4 Membandingkan nilai siluet dan Sum of Squares Error SSE modifikasi metode k-means OWA dengan metode klastering lainnya dalam klastering varietas padi. Hasil penelitian ini antara lain mendapatkan 8 variabel yang berpengaruh dalam klastering padi berdasarkan preferensi konsumen yaitu kadar amilosa suhu gelatinisasi persen beras kepala persen beras putih panjang bulir beras bentuk bulir beras pengapuran dan kadar protein. Tingkat akurasi modifikasi k-means OWA dalam klastering data set iris adalah 9667 yang jauh lebih baik daripada metode k-means yaitu 8933.Berdasarkan nilai siluet dan SSE maka jumlah klaster yang paling sesuai data set padi adalah 7 klaster dan modifikasi metode k-means OWA pada 945 0.8 paling baik dibandingkan metode k-means dan hierarchical clustering dalam klastering varietas padi karena nilai SSEnya paling kecil.


ABSTRACT

K-means clustering method based on Ordered Weighted Averaging OWA was developed by Cheng et al 2009 to resolve problem in classification using integrating k-means clustering and OWA. K-means clustering is a method of clustering and OWA is an aggregation operator. OWA was able to reduce the complexity of experimental data and help in representing sophisticated relationships between the criteria. Based on the original function of k-means and OWA algorithm used it is predicted that OWA-based k-means clustering Cheng et al 2009 works by modifying some of its stages. In this study it will be done by modification of OWA-based k-means clustering Cheng et al 2009 and applying it in the case of rice varieties clustering. This research aims to 1 collect variables related to consumer preferences in selecting a rice 2 apply OWA-based k-means clustering in clustering iris data sets for validation and the data sets of rice varieties 3 measure accuracy rate of OWA-based k-means clustering in the iris data sets 4 Compare the silhouette value and the Sum of Squares Error SSE of Modified OWA-based k-means clustering with other clustering methods in clustering rice varieties. Result showed that eight variables affecting rice varieties clustering based on consumer preferences namely amylose content gelatinization temperature percent of head rice white rice grain length grain shape chalkiness and protein content. The accuracy of OWA-based k-means clustering in clustering iris data sets is 96.67 which was better than k-means clustering method of 89.33. Based on the silhouette and the SSE value the number of clusters that best match the data set of rice varieties was 7 clusters. Modified OWA-based k-means clustering 945 0.8 compared to k-means clustering methods and hierarchical clustering was the best in clustering of rice varieties because it showed the smallest Sum of Squares Error.



Keywordsk-means clustering; OWA; Rice varieties.
 
Subject:  Analisis diskriminan
Contributor
  1. Ir. Budi Santosa, M.Sc., PhD
Date Create: 06/08/2010
Type: Text
Format: pdf
Language: Indonesian
Identifier: ITS-Master-3100010040556
Collection ID: 3100010040556
Call Number: RTI 519.53 Uly k


Source
Master Thesis of Industrial Engineering, RTI 519.53 Uly k, 2010

Coverage
ITS Community

Rights
Copyright @2010 by ITS Library. This publication is protected by copyright and per obtained from the ITS Library prior to any prohibited reproduction, storage in a re transmission in any form or by any means, electronic, mechanical, photocopying, reco For information regarding permission(s), write to ITS Library




[ Download - Open Access ]

  1.  ITS-Master-14804-2508201005-Abstract_id.pdf - 175 KB
  2.  ITS-Master-14804-2508201005-Abstract_en.pdf - 173 KB
  3.  ITS-Master-14804-2508201005-Conclusion.pdf - 176 KB




 Similar Document...




! ATTENTION !

To facilitate the activation process, please fill out the member application form correctly and completely

Registration activation of our members will process up to max 24 hours (confirm by email). Please wait patiently

POLLING

Bagaimana pendapat Anda tentang layanan repository kami ?

Bagus Sekali
Baik
Biasa
Jelek
Mengecewakan





You are connected from 10.199.6.2
using CCBot/2.0 (https://commoncrawl.org/faq/)



Copyright © ITS Library 2006 - 2021 - All rights reserved.
Dublin Core Metadata Initiative and OpenArchives Compatible
Developed by Hassan