Nina Begičević Ređep

A new method for strategic decision-making in higher education

Using the appropriate methodology for strategic decision-making in higher education is crucial to make effective decisions. In this paper, the analytic network process (ANP), one of the most suitable decision-making methods in terms of higher education issues, is presented and evaluated from the position of the user. After recognising some characteristics of the ANP that can be improved, the main objective of this research was to develop a new method based on the characteristics of the ANP and social network analysis (SNA). The research methodology used in this paper is the design science research process (DSRP), which is often used to design new artefacts, such as models, methods and methodologies. The main phases of this approach include problem identification, objectives of a solution, design and development, demonstration of the artefact, evaluation and dissemination. By using the DSRP, a new decision-making method is designed and proposed. The new method has two components that are based on the ANP and SNA. The first component is related to determining the importance of criteria with respect to the goal of decision-making. The second component is related to modelling influences/dependencies between criteria, and identifying criteria that strongly influence others, as well as criteria that others depend on. A measure that describes how strong a particular criterion is in terms of influences/dependencies is based on the centrality degree, one of the most fundamental centrality measures. In this paper, the new method, which was evaluated on several cases, is demonstrated with example of evaluating scientists, and a comparison of the new method’s results and the ANP method’s results is presented.

Data-driven decision-making in classification algorithm selection

Abstract The selection of the appropriate classification algorithm for a given data-set is an important and complex issue, full of research challenges. In this paper, we present a developed meta-analysis-based framework to improve decision-making in the selection of classification algorithms based on data-set characteristics. We study the effectiveness of our proposed framework with 32 data-sets. Three classification algorithms – neural networks, decision trees, and k-nearest neighbours – were trained and applied to data-sets with different characteristics, aiming to review the performance of algorithms in the presence of noise in the data, the interaction between features, as well as a small or a large ratio between the number of instances and the number of features. Our results show that feature noise is the most important predictor of the decision regarding the choice of the classification algorithm, and data-driven classification is found to be useful in this scenario.