Student academic performance using artificial intelligence

Abstract EN

This study is designed to find relationship between student's outcome of a particular course and their social backgrounds, previous achievements and the academic environments by using Artificial Intelligence.

Five hundred students from six departments at the Engineering College were participated in the research.

Firstly, the students have been tested before starting and after the completion of the course.

Then data about student's parental socio-economic status, tutors category; former high school scores, high school type and teaching languages have been collected.

The collected data has been pre-processed, cleaned, filtered, normalized and classified using Artificial Neural Network technique.

Initially, a 14 neuron neural network structure is proposed.

Then based on the classification and learning process, a modified model with 9 neurons is designed.

Each proposed methods are implemented and each is capable of performance predicting successfully.