Plant leaf diseases present a major challenge to agricultural productivity and global food security. The purpose of this study is to compare the novel decision tree method with the k-nearest neighbors in order to enhance the identification of plant leaf disease. To increase the detection rate of the current study, the decision tree technique with a sample size of 10 sets and a k-means clustering algorithm Considering a mean sample size of 10 sets a whole of 20 sets are contrasted. Using the ClinCalc software appliance under supervised learning, the mean accuracy of the current study was determined with a 95% confidence time frame, 0.5 alpha, 0.8 G-Power, and 0.02 beta values are obtained. Following this study, the k-nearest and novel decision trees both achieved 98.0% accuracy, and the accuracy of the k-nearest neighbors is 90.10%. After doing an independent samples examination of the T-test, p = 0.000 (p < 0.05) was discovered to be the significance value, indicating statistical significance. This study uses k-means clustering with the novel decision tree technique. It was discovered that the novel decision tree method outperformed the K-nearest neighbors once the current study trial was completed.