Comparison of Neural Network and Random Forest Classifier Performance on Dragon Fruit Disease

Measuring the performance of several classifiers in the modeling process, based on datasets with certain criteria, becomes important. It used to determine which classifier is more reliable for a particular task. Neural network will be compared its performance against Random Forest using dragon fruit datasets. Consisting of 41 images of healthy and sick fruit and leaf, it is divided into four classes, both classifiers were used in two comparing experiments. The validation used is 10-fold cross-validation. The results obtained are not much different from the prediction accuracy in the range of 70% to 82.9% for both classifiers.