Synthetic Learner Dataset for LEARN-RS Subject Recommendation Study

This dataset was generated for the development and evaluation of LEARN-RS, an interpretable machine learning recommender system for personalized subject recommendations in digital learning environments. The dataset contains synthetic records representing diverse learner profiles, including cognitive style attributes based on Kolb’s experiential learning dimensions, engagement metrics, prior academic performance, and demographic indicators. Each record is labeled with a recommended subject category (English, Chemistry, Art, or Project-Based Learning) derived from model predictions. The dataset includes the following fields: Concrete Experience (CE_Score): Numeric score reflecting experiential learning preference Reflective Observation (RO_Score): Numeric score reflecting reflective observation tendency Abstract Conceptualization (AC_Score): Numeric score reflecting conceptual learning inclination Active Experimentation (AE_Score): Numeric score reflecting active experimentation tendency Engagement_Score: Composite engagement metric Study_Habit_Consistency: Self-reported consistency of study habits Prior_Grades: Average prior academic performance Performance_Trend: Trend indicator of academic improvement or decline Subject_Familiarity: Self-assessed familiarity with subject domains Age: Learner age in years Recommended_Subject: Target label for recommendation This dataset was synthetically generated to simulate realistic learning behavior patterns while avoiding identifiable personal data. It is provided under a Creative Commons Attribution (CC BY 4.0) license to support reproducibility and further research in machine learning recommender systems.