Path: Top Journal Jurnal_Teknologi_&_Manajemen_Informatika 2007

Aplikasi jaringan syarat tiruan Metode Learning Vector Quantization dalam Pengenalan Huruf Alphabet

Jurnal Teknologi dan Manajemen Informatika, Volume 5, Nomor 2, Agustus 2007
Journal from JIPTUNMERPP / 2008-08-27 01:36:10
Oleh : Mohamad Yasin, Faculty of Information Technology Merdeka University Malang (myasin@plasa.com)
Dibuat : 2007-08-01, dengan file

Keyword : Artificial neural network, learning vector quantization method, alphabet presentation

The artificial intelligence does useful in helping human nowadays. One of its usages is to identify certain pattern presented. One example of patent recognition is alphabet identification. One branch of artificial intelligence is an artificial neural network. Within such neural network three are two kinds of learning process, i.e. observed- and not observed-learning. The distinction of these two learning processes is located in their output targets. Learning Vector Quantization methods is one method of artificial neural network with observed/controlled learning. In this method, neural network was trained using certain target data which was called load data and also training data. This data was calculated using the algorithm of Learning Vector Quantization, so that certain value of new load data was obtained. It was this score of new load data that was used for the computation of pattern recognition. The method of Learning Vector Quantization could be used in the identification of certain character. In this project, a specific program was created to recognize 26 alphabetic character and they were capital letters. These letters were in form of bitmap images that would be changed into that of binary matrices. The process of letter identification using any letters available in computer (font default windows) and the one using hand-made fonts. The result suggested a significant finding with the identification accuracy level of 75%.

Deskripsi Alternatif :

The artificial intelligence does useful in helping human nowadays. One of its usages is to identify certain pattern presented. One example of patent recognition is alphabet identification. One branch of artificial intelligence is an artificial neural network. Within such neural network three are two kinds of learning process, i.e. observed- and not observed-learning. The distinction of these two learning processes is located in their output targets. Learning Vector Quantization methods is one method of artificial neural network with observed/controlled learning. In this method, neural network was trained using certain target data which was called load data and also training data. This data was calculated using the algorithm of Learning Vector Quantization, so that certain value of new load data was obtained. It was this score of new load data that was used for the computation of pattern recognition. The method of Learning Vector Quantization could be used in the identification of certain character. In this project, a specific program was created to recognize 26 alphabetic character and they were capital letters. These letters were in form of bitmap images that would be changed into that of binary matrices. The process of letter identification using any letters available in computer (font default windows) and the one using hand-made fonts. The result suggested a significant finding with the identification accuracy level of 75%.

Copyrights : Copyright (c) 2008 by Digital Library Universitas Merdeka Malang. Verbatim copying and distribution of this entire article is permitted by author in any medium, provided this notice is preserved.

Beri Komentar ?#(0) | Bookmark

PropertiNilai Properti
ID PublisherJIPTUNMERPP
OrganisasiF
Nama KontakDra. Wiwik Supriyanti, SS
AlamatJl. Terusan Halimun 11 B
KotaMalang
DaerahJawa Timur
NegaraIndonesia
Telepon0341-563504
Fax0341-563504
E-mail Administratorperpus@unmer.ac.id
E-mail CKOwsupriyanti@yahoo.com

Print ...

Kontributor...

  • Editor: Wiwik Supriyanti, Dra. SS.