Abstract:

The number of students is increasing in Higher Education sector and to evaluate their academic performance is becoming challenging due to overwhelming of big data. Published research has taken necessary measures to visualize the learning performance to minimize the failure of students through traditional learning mode. Recently, the spread of pandemic COVID-19 led to cease of the physical academic session of students and imposed restriction on teaching via other pedagogies that include blended and online learning. The shift has motivated the study to conduct analysis on engineering students’ academic performance. This study used K-means data mining clustering technique to compare and analyse the engineering students’ academic performance in traditional, blended, and online learning methods. The academic and demographic attributes were used to identify the patterns for analysing the comparison of students’ performance in three learning methods using data of a private university in Malaysia. The results of k-means clustering analysis showed that the engineering students are doing better in online learning as compared to traditional and blended learning for the patterns investigated.

Keywords:

COVID-19, higher education, traditional learning, online learning, blended learning, attributes, data mining techniques.