(1) Simply use the same steps in the previous part and from the previous lab to implement SVM, kNN, and Logistic regression.

(2) Make sure to improve the accuracy of prediction by exploring and choosing the best number of features to include in the model.

(3) At each step you write in the code, please indicate what is the current task in the Machine learning workflow (write it as a comment before the code).

(4) Which one is better? SVM, k-NN or Logistic regression?

  

CCCY 431 - Machine Learning for Cybersecurity Task 2 - SVM Need apply SVM, k-nearest neighbor (k-NN) and logistic regression models on the "Zoo dataset", which has 17 features and one label "class_type". (value from 1 to 7 represent the 7 class types of Animal: Mammal, Bird, Reptile, Fish, Amphibian, Bug and Invertebrate). Your task is: (1) Simply use the same steps in the previous part and from the previous lab to implement SVM, KNN, and Logistic regression. (2) Make sure to improve the accuracy of prediction by exploring and choosing the best number of features to include in the model. (3) At each step you write in the code, please indicate what is the current task in the Machine learning workflow (write it as a comment before the code). (4) Which one is better? SVM, K-NN or Logistic regression? A Basic Machine Learning Workflow Representation Evaluation Optimization Choose A feature representation Type of classifier to use image pixels, with Choose What criterion distinguishes good vs. bad classifiers knearest neighbor classifier eg. % correct prediction on test Choose How to search for the settings parameters that give the best classifier. for this evaluation criterion eg. try a range of values for "" parameter in k-nearest neighbor clasifier Item Reading the dataset and make the split KNN model (create object, train, predict) each is 0.5 Logistic Regression model (create object, train, predict) each is 0.5 SVM model (create object, train, predict) each is 0.5 explore the best number of features and identify them. Redo at least one of the three model to prove the improvement State which method is better in the current experiment Mapping with Machine learning workflow Mark 0.5 0.5 0.5 1 1 1 0.25 0.25 5