ANALYSIS OF THE DEEP LEARNING APPROACH TO FIQH LEARNING IN FORMING STUDENTS' RELIGIOSITY (CASE STUDY AT MTS NEGERI 5 JAKARTA)

Fiqh learning in Islamic junior high schools is often trapped in a mechanistic memorization pattern (surface learning) so that it fails to internalize the spiritual values ??of worship into the real behavior of students. This study aims to analyze the application of the Deep Learning approach (mindful, meaningful, and joyful learning pillars) in Fiqh learning on obligatory prayer material, identify its contribution to the formation of five dimensions of religiosity, and map the supporting and inhibiting factors at MTs Negeri 5 Jakarta. The research method uses an embedded mixed approach (QUAL(quan)) with data collection through observation, in-depth interviews, documentation, and Likert-scale questionnaires to 16 grade VII students. The results show that the application of Deep Learning produces an average religiosity achievement of 85.75% (very strong category), with details of the achievement of the dimensions of faith (aqidah) of 88.75%, ihsan 87.81%, religious knowledge 85.94%, morals 83.75%, and worship 82.50%. The distribution of students' religiosity typology was dominated by the medium category (62.50%), followed by the high category (25.00%), and the low category (12.50%). The study's conclusion confirms that conscious, meaningful, and enjoyable classroom conditioning can transform pseudo-discipline into internal spiritual awareness (muraqabah) in students.