<![CDATA[In Indonesia, the rate of inaccurate major selection among first-year university students reaches 35%, primarily caused by external pressures without adequate self-assessment, leading to a 68% decrease in learning motivation and an increase in the dropout rate by 25%. This study aims to design a web-based expert system using a weighted forward-chaining algorithm to provide accurate major recommendations with a minimum accuracy of ≥80%, while also evaluating user acceptance levels. The research adopts a Research and Development (R&D) approach with a Model-View-Controller (MVC) architecture and a knowledge base comprising 50 decision rules integrating academic performance, RIASEC test results (30 items), non-academic experiences (15 items), and career aspirations (6 items). Validation involved 120 high school students (Grade XII) in West Kalimantan, employing a confusion matrix for accuracy testing and the System Usability Scale (SUS) for usability evaluation. Results show the system achieved 86% accuracy, 85% precision, 87% recall, with an average response time of 2.3 seconds. Expert validation yielded scores ranging from 4.0 to 4.4/5.0, while the SUS score reached 82.5±6.3 (“Excellent” category), with a 40% improvement in user trust through the explanation facility. The proposed expert system effectively provides accurate, adaptive, and user-accepted recommendations, contributing to reducing major mismatch and supporting the digital transformation of higher education.]]>
Copyrights © 2025
