DESIGNING COPTAS AS A COFFEE LEAF DISEASE DIAGNOSTIC APPLICATION BASED ON MACHINE LEARNING

Caku. Designing Coptas As A Coffee Leaf Disease Diagnostic Application Based On Machine Learning. Thesis. English Language Education Study Program of STKIP Persada Khatulistiwa. Advisor I: Sijono, M.Pd. Advisor II: Tuti, M.Pd. Keywords: machine learning, coffee leaf, disease diagnosis, mobile application COPTAS is a machine learning-based application designed to diagnose coffee leaf diseases such as rust, miner, and phoma, and identify healthy leaf conditions. In addition, COPTAS provides appropriate treatment recommendations based on the diagnosis results. The development of this application uses the ADDIE model, which consists of five stages: Analysis, Design, Development, Implementation, and Evaluation. This application aims to improve accuracy in the diagnosis of coffee leaf diseases while overcoming the limitations of manual methods that are time-consuming and prone to human error. Evaluation was conducted through two stages: internal testing by the development team and external testing through presentations in front of mentors and other teams at Bangkit Academy. Based on the evaluation results, COPTAS proved effective in analyzing coffee leaf diseases and providing accurate diagnosis results. Overall, the app functions well and is reliable. Further development is recommended, especially in improving the user interface and expanding the scope of disease types that can be diagnosed, so that the benefits for coffee farmers are optimized.