Learning Experiences Using Neural Networks and Support Vector Machine (SVM)

Abstract

This article is part of the global data mining framework, it addresses the theme of learning and classification, to identify the classes to which objects belong from using some descriptive parameters. They are particularly suited to the problem of automated decision-making. In this article we tried to implement three learning techniques, the Support Vector Machine (SVM), the Neural Networks and the Decision Trees. This application study aims to compare the results of these three techniques in terms of respecting the performance of the classification used for the contained objects in the data set ̳‘IRIS‘‘ based on the confusion matrix generated by the software weka, which is the tool used to carry out these learning experiences.