STEM-BASED BAYESIAN COMPUTATIONAL LEARNING MODEL-BCLM FOR EFFECTIVE LEARNING OF BAYESIAN STATISTICS

Abstract

This work contributes to the comprehension of Bayes’ theorem inclusive Bayesian probabilities and Bayesian inferencing within the framework of STEM (Science, Technology, Engineering, Arts, and Mathematics) and cognitive learning w.r.t Bloom’s taxonomy (BT). Bayes’ theorem is taken as a crucial statistical instrument employed in the development of intelligent systems and the management of risks, commonly utilized by engineers for tasks in machine learning and managerial decision-making. The fundamental concept behind Bayes’ theorem revolves around comprehending the degree of truth within the confines of an explicit perspective. This involves partitioning the entire sample space of possible evidence and utilizing the subset containing the relevant perspective to estimate the uncertainty of an event or the reliability of a model. However, it is often found difficult for students to understand Bayes’ theorem to the level of applying it to real-world problems. Considering this, the proposed learning method in this paper elucidated the acquisition of Bayes’ mathematical formulation by leveraging computational thinking, leading to the development of a computational model. The proposed model is named the Bayesian Computational Learning Model (BCLM). Subsequently, we have probed the utility of BCLM in the design and plan of learning activities, coherent to the STEM paradigm and BT cognitive learning hierarchy.