1. normalise the data using the z-score normalisation (For the purpose of this exercise, please use all data to compute the and required for the normalisation. Keep in mind, however, that in reality you have to use only the training data to compute and the input for the future predictions will typically not yet be available to you!) 2. Use the k-Nearest Neighbours method with k = 3 to predict the missing classes. Use the Euclidean norm in your distance calculations. Note: When making predictions, you need to transform the input data for the pre- dictions using the same mu and as for the transformation of the training data in (1).

Petal Length Petal width Class 1.7 virginica 5.8 1.5 0.3 setosa 4.9 2.1 virginica versicolor 4.6 1.3 0.3 setosa 1.4 6.2 2.4 1.4 4.5 5.8 2.4 versicolor versicolor versicolor ? ? ? 6.1 2.2 1.2 0.3 6.1 2.2 4.6 1.4 1.4 0.2 1.2 0.4 ? Petal Length Petal width Class 1.7 virginica 5.8 1.5 0.3 setosa 4.9 2.1 virginica versicolor 4.6 1.3 0.3 setosa 1.4 6.2 2.4 1.4 4.5 5.8 2.4 versicolor versicolor versicolor ? ? ? 6.1 2.2 1.2 0.3 6.1 2.2 4.6 1.4 1.4 0.2 1.2 0.4