Personalized Adaptive Learning Pathway System Using Reinforcement Learning, Knowledge Graphs, and Rule-Based Explainability

The paper develops a Personalized Adaptive Learning Pathway System (PALPS) which adaptively designs streamlined educational programs for compiler design courses. The system deals with three primary e-learning obstacles which include static delivery of educational material and inadequate personalized learning experiences as well as unclear recommendations to students. The system consolidates three fundamental elements including a multi-dimensional knowledge graph with relationship types exceeding twelve and a thorough learner profile and explainable rule-based recommendation algorithms. Development of the knowledge graph in Neo4j included 74 compiler design subjects which link with multi-dimensional associations to show conceptual associations that exceed basic prerequisite relations. The model takes learner cognitive state together with learning preference information and engagement patterns to produce expert learning pathways. Preliminary validation shows our approach improves path relevance by identifying cross-domain connections between topics that traditional hierarchical models miss. The rule-based layer enables users to see recommendation explanations which increases their trust in the system. The research fosters adaptive education by adding better knowledge structures while developing complete student models and explainable suggestion systems. The framework displays capabilities for use in technologies beyond compiler education which demand individualized learning approaches in structured technical fields.