Analisa Peningkatan Softskill dan Hardskill Siswa melalui Kurikulum Merdeka dengan Machine Learning

Abstract

The Merdeka Curriculum serves as a foundation for developing skills, both hard and soft, to prepare students to face the complex demands of society and the working world. However, assessments of improvements in soft and hard skills often lack consistency due to the subjective nature of evaluation by educators. This research aims to address these challenges by integrating Machine Learning (ML) methods into the assessment process. ML methods, including multiple linear regression, support vector machine (svm), and artificial neural network (ann), are implemented to help achieve more objective and consistent evaluations of students' progress in soft and hard skills. The research also involves comparing the accuracy of these three machine learning methods, resulting in 91.13%, 88.92%, and 91.26% accuracy, respectively. This research is expected to make a valuable contribution to understanding the effectiveness of the Merdeka Curriculum in relation to students' skill development. The comparison of accuracy of machine learning methods is also expected to provide important insights for educators, policymakers, and education stakeholders to support more informed and data-driven decision-making.