AI-Powered Student Dropout Prediction & Personalized Learning System

Abstract

As the technologies entered the new phase using artificial intelligence in many sectors and results in improvised performance by reducing time spent over simple analysis which is done manually, in this phase of evolution the modern education systems use many online tools for teaching and learning. Due to this, a large amount of students data is generated every day using these online learning platforms, assignments, tests and attendance records. Even-though this data is available, it is not utilized in proper way to support student at the right time. Due to this students results drops or they start thinking about leaving their course. In most of the institutions, student evaluations are still depends mostly on final exams and grades. This makes it difficult for institutions to notice early signs of students academic difficulty. For students who lose interest slowly or miss classes are identified only after their performance becomes poor. At that stage, it becomes harder for them to be recovered. To overcome this issue, the student risk prediction and specialized learning support system is proposed. The system regularly collects student attendance records, academic scores, learning activity, and financial data. By using this data , it calculates a simple risk score that groups students into low, medium, or high risk levels. Separate dashboard are available for students, teachers and parents to view their progress and receive frequent updates. This approach of tracking students helps in taking early action, improving student participation, and reducing the chances of dropout.