Klasifikasi Jenis Buah Durian dengan Metode K-Nearst Neighbor

Abstrak

Durian is a very famous fruit both in Indonesia and in the world, which is a native tropical fruit, where currently the classification process is still traditional or not yet computerized, namely based on the experience of farmers, seen from the color and shape, which sometimes still contains errors. in the classification process. This research aims to facilitate the process of classifying durian fruit using the k-nearest neighbor method. This system was built using MATLAB software, and uses a waterfall model in system development which consists of the analysis stage, design stage, coding stage and testing stage. The testing method uses a confusion matrix which is divided into two parts, namely training data and testing data. The results of this research show that the durian fruit type classification system using the k-nearest neighbor method with image processing using image processing, successfully recognized 224 of the 240 training data, with 16 data that were not recognized. After testing using holdout validation with 60 data, the system achieved a success rate of 93.3%.