<![CDATA[This study aims to describe deep learning-based co-curricular activities in developing students' scientific talents and attitudes at MI Tamrinussibyan Sumbersari Kayen, strategies for implementing the deep learning approach, as well as challenges and solutions in its implementation. Data were collected through interviews, observations, and documentation, with validation using source and method triangulation. Data analysis was conducted using the Miles and Huberman interactive analysis model which includes three main stages: data reduction, data presentation, and drawing conclusions or verification. The results of the study indicate that co-curricular activities such as scouting, dance, and the rebana club are able to develop students' talents and scientific attitudes. The strategy used in deep learning-based co-curricular activities is that teachers not only act as knowledge transmitters, but also as facilitators who encourage students to think critically, solve problems, and relate learning to real life. The challenges faced include limited facilities and infrastructure, lack of teacher training, and time management of activities. Solutions implemented include improving teacher training, more structured management of co-curricular activities, and strengthening school support for the implementation of deep learning. This research is expected to be a reference in the development of deep learning-based education in elementary education.]]>
Copyrights © 2026
