The quality of cocoa fruit is an important factor in supporting productivity and market value in the agricultural sector. The determination of superior cocoa fruit, which has so far been done manually, is still subjective and inefficient. This study aims to develop an expert system to classify superior cocoa fruit using the Fuzzy Logic method with the Mamdani approach, based on physical parameters such as length, width, weight and number of seeds. The research was conducted in Genggelang Village, Senara Hamlet, North Lombok, using 50 cocoa fruit samples from various varieties. The system was designed using the Python programming language and a MySQL database and tested through the black box method and performance evaluation using a confusion matrix. The classification results show that the system is capable of identifying superior cocoa fruits with an accuracy of 71.7%. The high recall value for the 'superior' class indicates good system sensitivity, although precision can still be improved. Misclassifications generally occur in physical parameters that lie on the borderline between categories. This study demonstrates that the fuzzy logic method is effective in enhancing the objectivity and efficiency of cocoa fruit classification. The developed system can serve as a potential tool to support decision-making in technology-based agriculture.
Copyrights © 2025