Learning and Application of Different Machine Learning Methods (KNN, SVM, Decision Tree) in Different Datasets

With the continuous development of artificial intelligence, machine learning has shown great ability in classification and regression problems. For example, logistic regression, SVM methods, and neural networks are widely used in credit evaluation, image recognition, and sales forecasting. In this paper, KNN method, SVM method and decision tree method are used for iris data set, aircraft delay data set and diabetes dataset was tested by linear regression and Lasso regression ridge regression. We can conclude that different machine learning methods have different sensitivities to the partitioning of the dataset (controlled by random_state and test-size). When selecting the model parameters, the influence of these factors on the model performance should be considered.