ON OPTIMAL ANALYSIS AND EVALUATION OF TIME RESPONSE IN E-LEARNING SYSTEMS (NEURAL NETWORKS APPROACH)

This piece of research addresses an interdisciplinary challenging issue concerned with dynamic evaluation of e-Learning systems' performance. That issue considers e-learners' time response (equivalently convergence time) requested for reach correctly assigned/desired output determined answer. Additionally, presented interdisciplinary work integrates neuronal, psychology, cognitive, and computer sciences into e-educational environment. That's in order to introduce analytical systematic study for the adopted e-learners' time response phenomenon.
In other words, performance modeling of the most complex biological neural system (human brain ) associated with e-learning process, have been considered as recent interdisciplinary trend adopted by educationalists in learning science incorporated Neuro-physiology, Psychology, and Cognitive science. Accordingly, Artificial Neural Networks (ANNS) models have been introduced as relevant discipline to investigate systematically mysteries of the basic two human brain functions (learning and memory).
Herein, response time has been specifically adopted as an appropriate candidate learning parameter applicable for reaching optimal analysis and evaluation of e-learning systems performance. The article pays a special attention towards realistic investigation of the e-learning response time phenomenon tightly coupled with students' brain function. Noting that evaluated dynamic time response parameter has been affected by the natural individual differences phenomenon of learners' brain role (number of neurons) while performing e-learning processes. Consequently, Artificial Neural Networks models have been adopted for realistic evaluation of timely dependant response till reaching desired correct output solution for any arbitrary question submitted during learning processes' examination. Interestingly, presented analysis and evaluation of time response mainly concerned with students' brain structure (synapses, axons, and dendrites) and function contribute to learning process at the macro-level. By more details, at any time instant; the status of synaptic connectivity pattern (vector) inside learner's brain supposed to be presented as timely dependent (changes) of weight vector. This status pattern expected to proceed spontaneously towards learner's correct output response (answer). Obviously, the number of contributing neurons (inside Learner's brain) has a significant effect on evaluated performance of any Learning process dynamics. After successful timely updating of dynamical state pattern (weight vector), pre-assigned achievement is accomplished in accordance with coincidence learning paradigm. Finally, other some interesting time response evaluation results introduced as inspired by intrinsic neuronal parameters (such as gain factor) after running of designed ANN simulation programs.