Viktors Zagorskis, Atis Kapenieks, and Aleksandrs Gorbunovs
Complex mathematical approaches exist in biological, social, and educational sciences, creating models to understand and explain cognition processes in human brain. Yet, the logged raw data is just an initial learners’ behavior footprint in Virtual Learning Environments. Exploratory Data analysis would help to deepen the understanding of cognition processes in students’ brain. The challenge is: the evaluation of usability of e-learning courses before the large-scale implementation. With this aim, we combine knowledge elements explored from logged learners behavior data and cognitive theories to formulate a computer model for Virtual Student’s evolution. We assume that some of the brains energetic expenses in learning and memorization are due to energy-extraction skills from cognitive abilities, knowledge, information, etc. We also assume that the Virtual Student model can perform energy flow modeling by extracting energy from the environmental learning objects and losing the power in a tedious learning process. The research shows that on a Virtual Student’s cognitive energy flow model based approach can potentially improve the model compliance with real student’s behavior model and can be applied to predictive analytics classification problems in both inexpensive small and large-scale applications.