<![CDATA[This study aims to analyze the implementation of deep learning in improving the descriptive text writing skills of grade VI students in Group 1, Bangkala District, Jeneponto Regency. Deep learning is seen as relevant to overcome students' low writing ability which is characterized by a lack of detail, weak text structure, and lack of ability to choose diction. This study uses a qualitative approach with the type of Class Action Research (PTK) of the Kemmis and McTaggart model which is carried out in one cycle consisting of planning, action, observation, and reflection stages. Data was collected through observation, interviews, documentation, as well as pretest and posttest. The results of the study showed a significant increase in description writing skills after the application of Deep Learning. The average pretest score of 58.40 with 16% completeness increased to 83.20 in the posttest with 88% completeness. An increase in N-gain of 0.62 indicates a moderate–high increase category. In addition to improved grades, students showed positive changes in learning behavior, characterized by increased participation, analytical skills, and quality of reflection on their writing. This study concludes that deep learning is effective in improving descriptive text writing skills while fostering students' critical thinking and metacognitive skills. This model is feasible to be applied as an alternative to Indonesian learning strategies in elementary schools.]]>
