Optimizing Deep Learning in Civics Learning: A Model Development Strategy to Improve Students' Democratic Literacy

Abstract

This study aims to optimize the application of Deep Learning models in Civics learning to strengthen students' democratic literacy. The study was conducted using a quantitative approach involving 46 respondents through purposive sampling, consisting of students, teachers, and education students. Data were collected through an online questionnaire and analyzed statistically. The questionnaire data showed a predominantly positive response (66.14%) to in-depth Civics learning, where students tended to actively understand concepts, hear them with real-life experiences, and participate in discussions. However, negative responses (33.6%) were still significant, caused by factors such as rote memorization habits, low motivation, or monotonous teacher-centered learning experiences, indicating that Civics learning does not fully medkkl student needs. Previous studies have shown that the use of AI can increase active participation, encourage critical reflection skills, and produce more analytical student responses, thus supporting the development of 21st-century skills. The findings of this study confirm that the integration of Deep Learning can be an effective strategy in Civics learning to create an adaptive, relevant, and engaging learning experience for students.