Neural Network Prediction of Suitability Course for Post PMR Students

This study aims to develop a Neural Network Model for predicting suitability course for post PMR students. Artificial Neural Network technique, using Multilayer Perceptron and backpropagation algorithm was employed in this case study. A total of 127 data sample of Form Four students of year 2003 fr...

Full description

Saved in:

Bibliographic Details
Main Author:	Tuck, Looi
Format:	Thesis
Language:	eng
Published:	2003
Subjects:	QA71-90 Instruments and machines
Online Access:	https://etd.uum.edu.my/1031/1/LOOI_TUCK.pdf
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my-uum-etd.1031
record_format	uketd_dc
spelling	my-uum-etd.10312013-07-24T12:10:05Z Neural Network Prediction of Suitability Course for Post PMR Students 2003-09-30 Tuck, Looi Sekolah Siswazah Graduate School QA71-90 Instruments and machines This study aims to develop a Neural Network Model for predicting suitability course for post PMR students. Artificial Neural Network technique, using Multilayer Perceptron and backpropagation algorithm was employed in this case study. A total of 127 data sample of Form Four students of year 2003 from SMK St Michael, Alor Star was trained using the above mentioned algorithm. The findings show that the best model composes of 10 nodes in input layer; 7 nodes in hidden layer and one node in output layer. A training percentage of correctness 81.04% and testing percentage of correctness 76.79 were achieved using this model. By identifying critical factors to the students’ suitability to the course, students and parents can make an informed decision. This project should be able to provide us with some insights into the type of pattern that exits in educational data. Therefore, Neural Network has great potential in educational planning. 2003-09 Thesis https://etd.uum.edu.my/1031/ https://etd.uum.edu.my/1031/1/LOOI_TUCK.pdf application/pdf eng validuser masters masters Universiti Utara Malaysia
institution	Universiti Utara Malaysia
collection	UUM ETD
language	eng
topic	QA71-90 Instruments and machines
spellingShingle	QA71-90 Instruments and machines Tuck, Looi Neural Network Prediction of Suitability Course for Post PMR Students
description	This study aims to develop a Neural Network Model for predicting suitability course for post PMR students. Artificial Neural Network technique, using Multilayer Perceptron and backpropagation algorithm was employed in this case study. A total of 127 data sample of Form Four students of year 2003 from SMK St Michael, Alor Star was trained using the above mentioned algorithm. The findings show that the best model composes of 10 nodes in input layer; 7 nodes in hidden layer and one node in output layer. A training percentage of correctness 81.04% and testing percentage of correctness 76.79 were achieved using this model. By identifying critical factors to the students’ suitability to the course, students and parents can make an informed decision. This project should be able to provide us with some insights into the type of pattern that exits in educational data. Therefore, Neural Network has great potential in educational planning.
format	Thesis
qualification_name	masters
qualification_level	Master's degree
author	Tuck, Looi
author_facet	Tuck, Looi
author_sort	Tuck, Looi
title	Neural Network Prediction of Suitability Course for Post PMR Students
title_short	Neural Network Prediction of Suitability Course for Post PMR Students
title_full	Neural Network Prediction of Suitability Course for Post PMR Students
title_fullStr	Neural Network Prediction of Suitability Course for Post PMR Students
title_full_unstemmed	Neural Network Prediction of Suitability Course for Post PMR Students
title_sort	neural network prediction of suitability course for post pmr students
granting_institution	Universiti Utara Malaysia
granting_department	Sekolah Siswazah
publishDate	2003
url	https://etd.uum.edu.my/1031/1/LOOI_TUCK.pdf
_version_	1747827046855213056

Neural Network Prediction of Suitability Course for Post PMR Students

Similar Items