Diagnostic Analysis Process of Forgettable Causal Relationship in Online Learning Behavior and Its Enabling Adaptive Enhancement Mechanism

Abstract

The interactive learning environment provides learners with a more autonomous and personalized learning mode. The learning behavior can be completely led by learners, and their participation and learning state will be constrained by various factors. The forgettable causal relationship is an important issue that relates various factors of dynamic and sustainable interactive learning process. For learning effectiveness, the forgettable causal relationship is not only a potential hazard, but also contains implicit data values. This study obtains the learning behavior instances of OULAD
with a data scale of 1.6PB, that covers 32593 learners, innovatively construct a novel diagnostic analysis process DAP-FCR with two important mechanisms ELBP and MLBP for forgettable casual relationship, and evaluate the learning trend through the unified calculation of attributes, features, and learning behavior routing. Sufficient comparative experiments have proved this diagnostic analysis process of DAP-FCR is feasible and reliable, the key statistic indicators Recall, Running Time, NDCG, Accuracy and Precision might be have been experimentally proven to be superior to other models currently available. Meanwhile, key values are mined to guide and improve learning behavior. Then the forgettable causal relationship might be adaptively transformed into learners' potential self-awareness and positive attitude about online learning, from "asking me to learn" to "I want to learn", and the interactive learning mode has become an "interesting thing", which has important practical significance for improving learning interest and developing learning preference, increasing learning attention and participation, and the flexible
enhancement mechanism of forgettable causal relationship is indirectly explored to optimize learning behavior routing and improving learning effectiveness.