Comparative Analysis to Determine the Best Accuracy of Classification Methods

The classification method is one of the methods of supervised learning and predictive learning. This method can be used to detect an object in the image presented, whether it is in accordance with the existing object in the training phase. There are several classification methods used, including Support Vector Machine (SVM), K-Nearest Neighbors (K-NN) and Decision Tree. To determine the accuracy in detecting these objects, it is necessary to measure the accuracy of each used classification method. The object that became simulation in this research was the object image of Guava and Pear fruit. Testing using confusion matrix. The results showed that the Support Vector Machine (SVM) method was able to detect with an accuracy of 98.09%. Then the K-Nearest Neighbors (K-NN) method with an accuracy of 98.06%, then the Decision Tree method with an accuracy of 97.57%. From the results of the accuracy test, it can be concluded that basically these three classification methods have high accuracy with a difference of 0.49% and the overall average accuracy of the classification of the three methods is 97.89%.