Data mining on ICT usage in an academic campus: A case study

Abstract

Nowadays, every higher education institution needs to assess the degree of utilisation of Information and Communication Technology (ICT) facilities installed in the campus. This study is based on the responses collected from the student and research communities via survey on ICT in the University of Burdwan. Data mining methodologies – Fuzzy-Rough Feature Selection (FRFS) to reduce the dimensions of survey dataset and subsequently different classification techniques such as J48, JRip, QuickRules Fuzzy-Rough rule induction, Fuzzy Nearest Neighbour, Fuzzy Rough Nearest Neighbour and Vaguely Quantified Nearest Neighbour (VQNN) are applied on this resultant dataset to build model for potential knowledge extraction. Fuzzy-Rough Feature Selection (FRFS) and then different classification algorithms are applied using WEKA 3.7.9 for analysis of the reduced dataset.