Prediction of Student’s Enrolment using Enhanced Extreme Learning Machines

The sustainability of educational institutions generally depends largely on strategic planning, both in terms of optimal allocation of resources/manpower and budgeting for financial aids/scholarships to incoming students. Hence, forecasting of student enrollment plays a vital role in making crucial decisions based on previous time-bound records. In this paper, forecasting of student enrollment is implemented using a neural network approach using extreme learning machines.Extreme learning machines are single hidden layer feedforward neural networks. They have extremely fast learning speed, good generalization performance, and universal approximation capabilities.The ELM model has a number of hyperparameters that is optimized for optimal performance –The hyperparameters in the ELM model such as lag size, number of hidden nodes, input weights and biases, and the type of activation function are optimized for optimal performance.