Using Machine Learning To Predict Student Course Registration Traffic In University

Abstract

Information systems are widely used in many industries and businesses. Educational institutions with complex business processes such as universities have been leveraging many information systems to aid their business, from managing students, courses, schedule, human resources, etc. This approach brings many benefits to the university and its students, yet it comes with challenges arising. One of the most notable issues for many universities comes from the course register system, which always gets slow and crashes during student registration sessions. There could be many reasons leading to the problem and many methods to address it. This thesis utilizes machine learning on the available data to create data models that can be used to assist university course managers and administrators of the system to make decisions when dealing with the mentioned problem. This is achieved by collecting data related to the course registration process from the database of the current system, specifically the database of International University. The collected data is then processed, analyzed, and used to train a model to give valuable insights for to help deal with the course registration congestion problem.