Student performance prediction with BPSO feature selection and CNN classifier

Educational Data Mining (EDM) is gaining great importance as a new interdisciplinary research field related to some other areas. It directly data mining (DM), the latter being fundamental part of knowledge discovery in databases (KDD). This growing more and contains hidden that could be very useful for users (both teachers students). convenient identify such form models, patterns, or any representation scheme allows better exploitation system. revealed tool achieve discovery, giving rise EDM. In this complex context, different techniques learning algorithms are usually used obtain best results. Recently educational systems adopting artificial intelligent systems, especially specific areas extracting relevant information, EDM, which integrates numerous support capture, processing, analysis these sets records. The main technique associated with EDM Machine Learning, has been decades processing contexts, but advent Big Data, there was an intensification application extract information from huge amount data. paper proposes student performance prediction using CNN (Convolution Neural Network) BPSO (Binary Particle Swarm Optimization) based feature selection method. study, classifiers made 2-class 5-class predictions. proposed system claims outperforming accuracy 96.6% various previous works well found majority attributes school activities compared on demographic socioeconomic characteristics.