Predicting Student's Performance using Data Mining Algorithm

Abstract

The term data mining refers to the practice of effectively extracting beneficial data from a large amount of data. Predicting a student's academic performance is the most complex and experimental study topic in educational data mining. Multiple factors have non-linear effects on performance, making this topic more appealing to researchers. researchers. This interest is enhanced by the increased availability of educational datasets, particularly in virtual education. There are several educational data mining surveys in the literature portion, we will only focus on student performance analysis and prediction. Data mining pursue a massive volume of dynamically created data for patterns and trends that are helpful and understandable to users. It can successfully utilize raw data generated by universities in examining hidden patterns and connections among the parameters that are used to estimate student performance and behaviour. Educational data mining bridges between the two disciplines: on the one hand is education and on the other in computer science. Educational actors (students, teachers, and administrators) have been benefitted as they are provided with the relevant information in which they have to act upon and thereby end up promoting quality-based innovations in this domain The main objectives of the system are to study existing data mining approaches in the educational domain and to analyze and compare the results of these approaches. We employed Support Vector Machine (SVM) and Naive Bayes (NB) to predict student performance in this paper.