PENGENALAN POLA KELAS BENANG MENGGUNAKAN METODE JARINGAN SYARAF TIRUAN BACKPROPAGATION

Giantara, Rangga Etyawan and Hidayatno, Achmad and Christiyono, Yuli (2011) PENGENALAN POLA KELAS BENANG MENGGUNAKAN METODE JARINGAN SYARAF TIRUAN BACKPROPAGATION. Undergraduate thesis, Diponegoro University.

[img]
Preview
PDF - Published Version
583Kb

Abstract

Pattern recognition represent a recognition to every data link (analog or digital), occurence and concept which can be differentiated. Form face, a desk, music couplet tone sequence, theme a symphony or rhyme, footstep made by particle plate fotografik, all the nya represent different type from pattern. Become, recognition a face, music couplet, painting, words from handwriting, diagnosis of disease from its symptom, and also recognition class of yarn all the problem of pattern recognition. Pattern recognition use this Artificial Neural Network method lifted in masterpiece write this. Artificial Neural Network represent one system process of information which design imitatedly is way job brain of human being in finishing a problem done process to learn through change of wight sinapsis. Artificial Neural Network able to recognize activity based only at past data. Past data will be learned by artificial neural network so that have ability to give decision to data which have never been learned. This method is weared to pattern recognition quality of yarn. Program used to do pattern recognition class of yarn with artificial neural network is Matlab version R2009a. Pursuant to examination research result of data entirety with variation hidden layer (hidden layer = 1, 2, 3, 4, 5, 6 and also 7 owning mean mount recognition equal to 82,9%. With highest recognition equal to 98,3% ( hidden layer = 6 ), while recognition lowest equal to 40,5% (hidden layer = 5 ). Side of this other, recognition level with the Artificial Nerve Network (JST) method simulation result with variation hidden layer measure up to erratic. So that, to obtain get network with variation hidden layer need training which is more amount. Key Words : Pattern recognition, Artificial Neural Network, Backpropagation, Hidden Layer

Item Type:Thesis (Undergraduate)
Subjects:T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions:Faculty of Engineering > Department of Electrical Engineering
Faculty of Engineering > Department of Electrical Engineering
ID Code:32059
Deposited By:INVALID USER
Deposited On:20 Dec 2011 13:23
Last Modified:20 Dec 2011 13:23

Repository Staff Only: item control page