Innovation: Predicting potential applicants who are probable to take admission is an essential aspect of a college/university. Our research will help the college/universities to increase their efficiency in terms of admissions and will attempt to reduce the effort and cost of the admission process. Machine Learning algorithms can change the face of modern marketing methods to be used by the college/universities and help them in identification of students based on various parameters such as 10th percentage, 12th percentage, JEE All India Rank, opted for coaching, availability of backup options, distinct extracurricular activities, reasons to apply in this college, yearly family income etc. In this paper, we have performed a comparative analysis by using Logistic Regression, Neural Networks, and Light Gradient Boosting Machine (LightGBM). These machine learning methods are used to carry out the prediction of the potential applicants who would take admission. LightGBM algorithm is used to study, which belongs to ensemble learning. Finally, the algorithm achieved an accuracy rate of 98.5%.
Published : IEEE Conference https://ieeexplore.ieee.org/document/9071525