Operationalizing Academic Performance in a Rule-Based KLSI-informed Decision Support System for Academic Specialization

Abstract

Decisions regarding academic specializations require detailed and transparent representations of learners. This study developed and evaluated a rule-based decision support system that transforms routinely collected academic performance data into two distinct analytical outputs: learner profiles informed by the Kolb Learning Style Inventory (KLSI) and curriculum-focused specialization rankings. The requirements were obtained through structured interviews with five curriculum-related informants from four public senior high schools in Malang, Indonesia. Subject achievement was deductively mapped to Concrete Experience (CE), Abstract Conceptualization (AC), Active Experimentation (AE), and Reflective Observation (RO) orientations as a theory-based operational knowledge representation. Specialization scores were calculated independently using predefined subject sets and ranked to identify the top two options for each participant. The evaluation included 12 specification-based test cases, 25 Black Box items, the System Usability Scale (SUS) administered to 40 Grade XI students, and descriptive analysis of 36 complete saved records. All specification cases matched manually derived outputs, functional compliance reached 92.0%, and the mean SUS score was 74.19. Diverger was the most frequently inferred profile (38.9%), while Social Humanities was the most common top-ranked specialization (44.4%). The system provides clear, traceable academic decision support without claiming psychometric equivalence to the traditional KLSI for use in school-counseling contexts.