Higher Education Units Assessment Based on Data Envelopment Analysis and Clustering Techniques

Higher education plays a vital role in community development and provides indicators of national strength in all aspects of life. Moreover, it is able to achieve the desired economic growth. Therefore, measurement performance of decision-making units is a vital process in this issue. Higher education institutions are multi-input and output institutions and this type of institutions is difficult to be evaluated using traditional economic methods. This study aims to evaluate the efficiency and quality of higher education institutions based on Data Envelopment Analysis (DEA). Furthermore, we also aim to find solutions and proposals for universities with low efficiency values (or inefficient) by identifying weaknesses in the resources of these universities.The DEA assumes that all Decision Making Units (DMUs) are homogenous in their environments while the DEA process is not enough to compare universities’ performances. So, our work proposes to use an unsupervised data mining technique such as kmeans algorithm to group universities with similar characteristics. Then, the DEA is utilized for each cluster separately. The result shows a better improvements and a fair comparison of performance between universities.