Document Details

Document Type : Thesis 
Document Title :
Mining E-Learning Environments to Enhance the Learning Process
استخدام تقنية التنقيب عن البيانات لتطوير العملية التعليمية في نظم التعليم عن بعد
 
Subject : Faculty of Computing and Information Technology 
Document Language : Arabic 
Abstract : Many e-learning systems have been developed and are in use around the world. Learning Content Management Systems (LCMS) store lots of data such as students' profiles and students' activities and interactions with the system, in addition to courses contents. This makes it almost impossible to manually analyze the data for valuable decision-making. This brings the need for automating the analysis of such data to reach quality decision making. There are many ways to do automated analysis on huge databanks; one of them is data-mining. By mining the LCMS database we can get a wealth of results. In this thesis, the focus was on how mining techniques can enhance the overall qualities of learning processes. Specifically, data mining is used to predict students performance based on their usage of the system, which is saved in the system's database. Attributes of performance are selected based on their direct impact on the quality of learning process based on archived data of the previous course intakes. Student performance prediction in e-learning environments is not apparent as it is in face-to-face traditional education, where the instructor can take direct feedback from the students. In e-learning summative methods, student performance prediction is based on the student’s results of the exams; however, formative prediction depends on many learning inputs, and hence, is difficult in tradition education. A formative prediction is more achievable in e-learning environment by mining archived data about the student usage and his/her interaction with the LCMS. In this thesis, we developed a model, a framework and a prototype for Student Performance Prediction and Advising (SPP&A), which employs data mining techniques. The system takes the role of an academic advisor who analyzes and predicts students performance and advises students and instructors on usage strategies for performance improvements. SPP&A used the known C4.5 algorithm for data classification. The C4.5 algorithm builds a decision tree based on the most critical LCMS usage attributes. This tree is used for both predicting and advising. Providing students with performance improving strategies would not only improve the learning results but would also elevate the student's moral for inherent learning results improvements. 
Supervisor : Dr. Shehab Ahmad Gamalel-Din 
Thesis Type : Master Thesis 
Publishing Year : 1431 AH
2010 AD 
Co-Supervisor : Dr. Adnan Moustafa Al-Bar 
Added Date : Saturday, July 24, 2010 

Researchers

Researcher Name (Arabic)Researcher Name (English)Researcher TypeDr GradeEmail
ناهد عبد العزيز العويضيAl-Owadi, Nahed AbdulazizResearcherMaster 

Files

File NameTypeDescription
 27490.pdf pdf 

Back To Researches Page