Development of an AI-Based Assessment Analytics Model for Transformative Teacher Professional Education

Transformative Teacher Professional Education (PPG) requires an assessment system that not only produces scores, but is also able to map competency achievements, identify risks, and provide data-driven follow-up recommendations. This research aims to develop an artificial intelligence-based assessment analytics model to support the monitoring of academic development and intervention of PPG UIN Syarif Hidayatullah Jakarta students. The research uses a Research and Development approach with the ADDIE model which includes the Analyze, Design, Develop, Implement, and Evaluate stages. Data was obtained through needs analysis, document review, interviews, observations, expert validation, system tests, user questionnaires, and assessment data in the form of tryout results and reflective essay responses. The results of the analysis showed that 91.6% of students needed automatic learning recommendations and 94.1% needed individual competency weakness mapping. The product developed is a web-based platform that contains a tryout dashboard, reflective essay analysis, risk mapping, essay forms, an administration panel, data filtering, and report export. The results of expert validation obtained an average score of 4.94 with the very feasible category, while the implementation results obtained an average score of 4.91 with the very feasible category. This model is considered very practical, operationally effective, and beneficial for supervisors and PPG managers. These findings suggest that AI-based assessment analytics have the potential to strengthen diagnostic assessments, achievement monitoring, and data-driven academic interventions, although their impact on graduation improvement still requires longitudinal testing.