<![CDATA[Many pesantren (Islamic boarding schools) still rely on teacher-centered instruction, and available digital technology is rarely designed to support deep learning. This study developed and evaluated the Machine Learning Platform-Enhanced Deep Learning Model (MLP-DLM), which combines deep learning pedagogy (6C), machine learning platform features, and core pesantren values. Following the ADDIE model, the product was validated by three experts, field-tested with 30 students, and evaluated in a quasi-experimental pretest-posttest control group design involving 60 students in two intact classes. Analyses used t-tests, N-gain, ANCOVA, and Cohen's d. The model scored 4.72 out of 5.00 in expert validation and 87.5% in usability, with response scores of 89.5% from students and 92.8% from teachers. With equivalent pretest scores, the experimental group improved significantly more than the control group in conceptual understanding (d = 1.83), digital literacy (d = 1.86), and higher-order cognitive engagement (d = 1.85). N-gain was moderate in the experimental group and low in the control group. ANCOVA supported these differences, although for digital literacy and higher-order cognitive engagement it counts only as supporting evidence. MLP-DLM is valid and practical and shows promise in this context. It should be replicated in more pesantren with longer interventions before wider adoption.]]>
Copyrights © 2026
