Abstract: The institutions of higher education now consider a key aspect in the development of educational reality in Iraq, and the focus of Iraqi universities and higher education institutions is nowadays, directed to enhance the learning environment for better learning opportunity of their students. With the information and communication technology era, we have a greater chance to make a whole new learning generation by creating intelligent electronic learning platforms that have a great impact on advancing students' educational level. In this research, we use one of the most well-known yet important artificial intelligence methods; one of them is called Educational Data Mining (EDM). That will be directed to extract hidden knowledge based on a pre-determined dataset from each student's learning behaviour inside a proposed electronic learning platform which is designed especially for higher education studies, the Department of Computer Science. Data repository contains all students' information. For example, age, country, university, previous occupation, and their grades obtained during the course, their learning behaviour, what course they choose, how many times did they re-take the exam until they pass, how many of them complete the course and how many of them have the course drooped. All these data are extracted after the proposed platform has been launched to a public website and used by actual students of higher education studies, then their learning data are kept in electronic learning repository to be used later to discover the most influencing factors about their learning process. For that purpose, two main questionaries' have been taken place, first one is taken prior the platform launching, and the second one taken after the platform launching and using by the students, the audience of the questionnaire were 100 students and faculty members in the institution of higher education, after the learning data have been accumulated, we use Multi Layer Perceptron (MLP) algorithm to evaluate the proposed e-learning platform performance to predict the most influencing factors in evaluating the proposed e-learning platform.

Keywords: Higher Education, Information Technology, Artificial Intelligence, Educational Data Mining (EDM), E-Learning Platform, Learning Behaviour, Multilayer Perceptron Algorithm (MLP)

| DOI: 10.17148/IARJSET.2019.6514