2. Predict the class of the given new input (3,4,6) using the 2-NN classifier, based on the training data (6 instances) given below. You will use the Euclidean distance to calculate the distance. Similarity (closeness, nearness) value will be the inverse of the distance. e.g.) if the distance is 3, then the similarity(closenessearness) value is 1/3 Justify your predicted outcome for the new input (3,4,6) by showing the distance (or similarity) values. Calculate the numbers rounded off to the second digit after the decimal point. (3,5, 8) => class: B (3, 2, 6) => class: A (2, 5, 10) => class: A (4,6,5) -> Class: B (4, 3, 7) => class: A (5, 4, 2) > class: B (3,4,6) Use the Boosting classification algorithm to classify the same training instances of the question (2), and to predict the outcome of the same new input (3,4,6). The training data file "data.csv" has been provided on eClass Iteration number: 3 Base classifier : Decision Tree (Use WEKA with the default option setting) The weight value for an incorrectly predicted instance will be doubled in the next phase. When you run WEKA, you have to choose 'Use training set' as the Test options instead of "Cross-validation'. 'Cross-validation' sets aside a portion of the training data during training. But in this assignment, you have to use the whole training set during the training phase. You are required to (1) write the training data with the weight values, in each iteration (2) write the training error rate of each classifier (error rate = (the number of incorrectly predicted instances)/(the number of all instances)) (3) provide the final prediction outcome for the new input (4) justify the final prediction outcome 5 A B Featurel IFeature2 3 3 2 4 4 5 u N D Feature3 Class 5 8 B 2 6 A 5 10 A 6 5 B 3 7A 4 2 B 2. Predict the class of the given new input (3,4,6) using the 2-NN classifier, based on the training data (6 instances) given below. You will use the Euclidean distance to calculate the distance. Similarity (closeness, nearness) value will be the inverse of the distance. e.g.) if the distance is 3, then the similarity(closenessearness) value is 1/3 Justify your predicted outcome for the new input (3,4,6) by showing the distance (or similarity) values. Calculate the numbers rounded off to the second digit after the decimal point. (3,5, 8) => class: B (3, 2, 6) => class: A (2, 5, 10) => class: A (4,6,5) -> Class: B (4, 3, 7) => class: A (5, 4, 2) > class: B (3,4,6) Use the Boosting classification algorithm to classify the same training instances of the question (2), and to predict the outcome of the same new input (3,4,6). The training data file "data.csv" has been provided on eClass Iteration number: 3 Base classifier : Decision Tree (Use WEKA with the default option setting) The weight value for an incorrectly predicted instance will be doubled in the next phase. When you run WEKA, you have to choose 'Use training set' as the Test options instead of "Cross-validation'. 'Cross-validation' sets aside a portion of the training data during training. But in this assignment, you have to use the whole training set during the training phase. You are required to (1) write the training data with the weight values, in each iteration (2) write the training error rate of each classifier (error rate = (the number of incorrectly predicted instances)/(the number of all instances)) (3) provide the final prediction outcome for the new input (4) justify the final prediction outcome 5 A B Featurel IFeature2 3 3 2 4 4 5 u N D Feature3 Class 5 8 B 2 6 A 5 10 A 6 5 B 3 7A 4 2 B