Document Type |
: |
Thesis |
Document Title |
: |
STUDENT DROPOUT PREDICTION IN MOOC’S USING DEEP LEARNING تنبؤ انسحاب الطلبة من الدورات المفتوحة على شبكة الإنترنت باستخدام تعلم العميق |
Subject |
: |
Faculty of Computing and Information Technology |
Document Language |
: |
Arabic |
Abstract |
: |
Massive Open Online Courses (MOOCs) have produced creativity, enthusiasm, involvement and various discussions from different contributors across the educational field. The enrollment in the MOOCs technology does not demand specific criteria, such as acquiring high intelligence within this area. Rather, this form of technology facilitates the enrollment in enormous online classes with the absence of any actual physical attendance. However, and despite such facilitation, the MOOCs does not get high levels of completion rates. Whilst considerable number of people apply for an online course, just few of them who accomplish the entire course. This dissertation intends to analyze MOOCs from a distinctive dataset that was provided from KDDCUP Competition 2015. The employment of machine learning mechanisms considers a key aspect for such analysis in order to categories contributors as either dropouts or completers. The engagement with different modalities of mechanisms, including Logistic Regression, Support Vector Machines (SVM), Gradient Boosting, Naïve Bayes, Random Forrest and Decision Trees are examples of such techniques. In addition, the utilization of deep learning models, such as the Deep Learning sample constructed from scratch and the Multi-Layer Perceptron can assist the analysis. Moreover, the dissertation aims to confirm to what extent various domains and systems could be analyze using a general process. Finally, an evaluation of the characteristics has been used in the analysis will be demonstrated |
Supervisor |
: |
Dr. Mu'adhim Siddiqui |
Thesis Type |
: |
Master Thesis |
Publishing Year |
: |
1440 AH
2019 AD |
Added Date |
: |
Tuesday, June 25, 2019 |
|
Researchers
فهد مسعود سيد | Sayed, Fahad Massoud | Researcher | Master | |
|