Multi layer feed forward artificial neural network for learning styles identification

Accommodating learning styles in adaptive educational hypermedia system (AEHS) may lead to an increased effectiveness and efficiency of the learning process as well as teacher and learner satisfaction. The premise is that a fact that learning in the classroom is less efficient, when teachers will no...

Full description

Saved in:
Bibliographic Details
Main Author: Bayasut, Bilal Luqman
Format: Thesis
Language:English
English
Published: 2015
Subjects:
Online Access:http://eprints.utem.edu.my/id/eprint/16821/1/Multi%20Layer%20Feed%20Forward%20Artificial%20Neural%20Network%20For%20Learning%20Styles%20Identification.pdf
http://eprints.utem.edu.my/id/eprint/16821/2/Multi%20layer%20feed%20forward%20artificial%20neural%20network%20for%20learning%20styles%20identification.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Accommodating learning styles in adaptive educational hypermedia system (AEHS) may lead to an increased effectiveness and efficiency of the learning process as well as teacher and learner satisfaction. The premise is that a fact that learning in the classroom is less efficient, when teachers will not be able to get insight of each of the student’s learning style; hence, they won't be able to adapt their teaching strategies to match with the student’s learning style. In order to get an insight of the student’s learning style in AEHS, the system must be able to recognize the learning styles of the students. Current methods for recognizing learning styles are less efficient, where questionnaires will lead to tedium and disturbance at learning processes. Thus, this study developed the learning styles based AEHS that utilized Multi Layer Feed-Forward Artificial Neural Network (MLFF) which was used to identify student’s learning styles in real-time. The automatic and real-time learning styles identification was done by analyzing the student’s browsing behavior while they are learning through the proposed AEHS. The system then adaptively presents the learning content that matches with the students’ learning styles by the means of fragment sorting and adaptive annotation technique. At the end of the study, the data triangulation was done to test if incorporating learning styles in learning environments can impact the student achievement. It was done by asking the student to answer the mini quiz after they were using the proposed AEHS with adaptive feature was activated. This study also focused on analysis of the existence of the relationship between the frequencies of students’ click on learning components with their staying time on those particular learning components. The result showed that the proposed MLFF performed well in identifying the students’ learning styles in real-time. Moreover, the analyzed student’s browsing behavior revealed that there was a relationship between the frequencies of the students’ click on learning components with their staying time on those particular components. Furthermore, after the student’s learnt through the proposed AEHS with adaptive feature activated and answered the mini quiz result; most of them could achieve the perfect score. In this case, the mini quiz result showed that incorporating learning styles into learning environment may affect and increase student’s achievements.