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


Taufik Rachmanu


Moh. Fandika Aqsa


Tondo Indra Nyata


Ansi Aflacha Putri


Aprillia Tri Wulansari


Dewi Eka Agustina




! 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

Still Confuse?
Please read our User Guide

Keyword
Mode
Expanded Search (for Free text search only)
 
OR You May Search Our Knowledges Using Google Customize Search :

Custom Search

ITS » Master Theses » Statistika - S2
Posted by ida at 26/12/2006 15:44:51  •  33874 Views


ANALISIS DISKRIMINAN MULTIVARIATE DENGAN METODE ARTIFICIAL NEURAL NETWORK(ANN)

MULTIVARIATE DISCRIMINANT ANALYSIS USING ARTIFICIAL NEURAL NETWORK (ANN) METHODS"

Created by :
SRIHARINI 



SubjectMatematika Teknik
Keyworddiscriminant analysis
artificial neural network
Misclassified

Description:

Analisis diskriminan merupakan salah satu teknik multivariate yang difokuskan pada pemisaban obyek (pengamatan). Tujuan dari analisis ini adalab menggambarkan bentuk yang berbeda dari beberapa populasi yang telali diketabui, sedemikian hingga populasi tersebut dapat terpisab dengan baik. Akan tetapi penggunaan teknik ini sering menimbulkan masalah jika data multivariate tidak berdistribusi nonnal atau tidak memenuhi asuinsi nonnal. Salah satu cara untuk menyelesaikan masalah ini adalah dengan pendekatan neural network yang didasari dari fungsi kepadatan peluang secara nonparametrik. Dalam penelitian ini akan dipelajari tentang analisis diskriminan dengan menggunakan pendekatan artificial neural network (ANN), dimana kelebihan ANN adalah bahwa fungsi yang digunakan tidak linear, mempunyai ketelitian yang tinggi serta tidak mempunyai model (nonparametrik), sehingga dengan pendekatan ini tidak diperlukan lagi asumsi dari data multivariate yang berdistribusi nonnal. Hasil analisis data dengan kedua metode, terlihat bahwa analisis diskriminan ANN pada data 3 cat merek "X" dengan 14 hidden node (satu hidden layer) akan memberikan hasil yang lebih baik dibandingkan hasil dari analisis diskriminan linier. Hal ini terlihat jelas dari kesalahan pengelompokannya. ANN tidak mengbasilkan kesalahan pengelompokan (kesalalian 0%), sedangkan kesalalian pengelompokan analisis diskriminan linier sebesar 7,4%. Dengan MSE (0,0396941) dari ANN lebih kecil dari pada MSE (0,1105894) analisis diskriminan linier.


Alt. Description

Discriminant analysis is one of many developed multivariate methods that pointed at separating obyect (obsevation). Direction of this analysis is discribe the forms difference of several known population is able to separate well. However this methods often having many problems if multivariate data is unable to distributed normally or uncomplete assumption normally one of the methods to solve this problem is using neural network approaching that based on function of probability in nonparametric. This research will learn about discriminant analysis using ANN probability, where superiority of ANN is using nonlinear function, high accuracy and nonparametrics. So that approaching do not need assumption by multivariate data that distributed normally. The result of data analysis using both methods above are indicate that ANN discriminant analysis with 14 hidden node (one hidden layer) will give more better result than linear discriminant analysis that visible at the Misclassified observations. ANN don't give Misclassified observations 0%), even linear discriminant is 7,4%. with MSE (0,0396941) is less than MSE of linear discriminant analysis (0,1105894).

Contributor:
  1. Drs. H. Nur Iriawan, Mikom, Ph.D
    Ir. Anik Djuraidah, MS
Date Create:20/12/2006
Type:Text
Format:pdf ; 42 pages
Language:Indonesian
Identifier:ITS-Master-3100003018397
Collection ID:3100002014841
Call Number:519.535 Sri a


Source :
Theses Statistics Engineering RTSt 519.535 Sri a, 2001

Coverage :
ITS Community

Rights :
Copyright @2005 by ITS Library. This publication is protected by copyright and permission should be obtained from the ITS Library prior to any prohibited reproduction, storage in a retrievel system, or transmission in any form or by any means, electronic, mechanical, photocopying, recording, or likewise. For information regarding permission(s), write to ITS Library


Publication URL :
http://digilib.its.ac.id/analisis-diskriminan-multivariate-dengan-metode-artificial-neural-networkann-279.html




[ Free Download - Free for All ]

...No Files...

[ FullText Content - Please, register first ]

  1. ITS-Master-279-1399201001-its-judul.pdf - 268 KB
  2. ITS-Master-279-1399201001-its-bab1.pdf - 94 KB
  3. ITS-Master-279-1399201001-its-bab2.pdf - 434 KB
  4. ITS-Master-279-1399201001-its-bab3.pdf - 163 KB
  5. ITS-Master-279-1399201001-its-bab4.pdf - 329 KB
  6. ITS-Master-279-1399201001-its-babkesimp.pdf - 165 KB
  7. ITS-Master-279-1399201001-its-lampiran.pdf - 773 KB



Give your Comment here...


 10 Similar Document...



 10 Related Document...






HELP US !
You can help us to define the exact keyword for this document by clicking the link below :

Misclassified , analysis , artificial , artificial neural network , discriminant , discriminant analysis , network , neural



POLLING

Bagaimana pendapat Anda tentang layanan repository kami ?

Bagus Sekali
Baik
Biasa
Jelek
Mengecewakan




53564716


Visitors Today : 1
Total Visitor : 2515801

Hits Today : 14327
Total Hits : 53564716

Visitors Online: 1


Calculated since
16 May 2012

You are connected from 10.199.13.5
using CCBot/2.0 (http://commoncrawl.org/faq/)


ITS Digital Repository Feeds



Copyright © ITS Library 2006 - 2014 - All rights reserved.
Dublin Core Metadata Initiative and OpenArchives Compatible
Developed by Hassan