A proposed framework in an intelligent recommender system for the college student

Abstract

Abstract This article aims to proposed framework an Intelligent Recommender System (IRS) for students in higher education institutions. This conceptual framework includes problems in predicting student performance, the possibility of graduating on time, and recommends choosing subjects according to performance, and career interests, which are useful for assisting pedagogical interventions in future student development. The success in the development and implementation of the proposed IRS framework is inseparable from using data mining and machine learning techniques in predicting and providing recommendations. Data analysis consisted of clustering techniques, association rules, and classification using Support Vector Machine (SVM), Naïve Bayes, and k-Nearest Neighbour (k-NN). These techniques are used to solve problems related to students and to provide appropriate recommendations. The result is an IRS conceptual framework for the college student that can be used as smart agents to provide student guidance and suggestions to support the process of education in higher education.