Deep learning for equity in education

Safa Hamreras

Safa Hamreras

Skikda, Skikda Province

This project predicts the students performance based on their demographic information. ...learn more

Project status: Published/In Market

Artificial Intelligence

Groups
SkaiLab

Overview / Usage

Education is the only path leading to a successful professional career, nevertheless, in this modern era, the worldwide population still suffers from the lack of equal opportunities to get a high quality education, or even accessing the desired study field. Education inequity could be observed between people based on their region, wealth, or gender. This project aims to solve this by using a deep neural network to predict students performance and show how education inequity is depriving skilled people of following their dream career and having a positive impact on their society.

Methodology / Approach

A deep neural network (DNN) is trained on Open University Learning Analytics (OULA) dataset to predict students final result in their study fields. The goodness of the approach is measured in terms of classification accuracy, this latter equals 0.78, after 50 epochs of training.

Technologies Used

Keras, Tensorflow, Google Colab, Tesla T4 GPU.

Collaborators

1 Result

1 Result

Comments (9)