Predicting the impact of different Variables on Students Academic Performance using Artificial Intelligence

Abstract- Undergraduate student’s academic performance is a long standing issue in higher education and a great deal of research over the past many years. At the end of each semester, students ’ result will be analyzed in order to evaluate students ’ academic performance. Thus there is a need for the prediction models which can effectively predict the student’s academic performance based on some relevant attributes which can be useful to guide the students with low academic record initially. Most of the existing research on the performance prediction uses various techniques such as classification, decision tree etc as a prediction methodology which basically classify the students rather than predicting the continuous grades of the students. In this paper, an attempt will be made to apply artificial neural network which is trained using backpropogation algorithm in predicting the academic performance of the students, which has some noticeable benefits over the other suggested prediction models in the educational domain. The benefits of the proposed system include: The more accuracy in prediction and can provide the continuous as well as classified discrete output