Decision Support for Managers: Using Algorithms to Optimize the Student Management Decision-Making Process

Distant learning revealed the requirement for evolution in education to adapt to social distinctions and modern technology. Teachers are assessed on their ability to teach as well as motivated and their students are influenced by a variety of factors. Enhancing comprehension and boosting student education and educational success are receiving more attention. To strengthen the processes of decision-making by streamlining data mining tasks, this research proposes an algorithmic-driven decision support system (Algo-DSS) technique. Initially, we gathered a graduate-level academic sample with a variety of attributes. Uniform Manifold Approximation and Projection (UMAP) and Principal Component analysis (PCA) techniques are used to extract the particular characteristics regarding dimensionality reduction. Moreover, the suggested structure for the feature extraction procedures encompasses the clustering methodology. This paper presents the unique swarm-inspired dynamic catboost (SIDCatBoost) approach for enhanced classification performance. The effective feature subsets are chosen using the swarm approach which is known as rock hyraxes optimization (RHO) and then fed into the catboost technique. Employing the suggested framework, the issue of mining academic information has been carefully investigated with the aim of evaluating student performance. In addition to demonstrating a significant boost in performance through the use of a technique for choosing discriminative data attributes, the task of classification was completed with excellent results. Algo-DSS has the potential to be a helpful tool in educational settings, especially for enhancing processes of decision-making, as demonstrated by the conducted study.