2. In this example we attempt to build a neural network that clusters iris flowers into natural classes, so that similar flowers are grouped together. Each iris flower is described by four features:

(a) Sepal length in cm

(b) Sepal width in cm

(c) Petal length in cm (d) Petal width in cm

This is a clustering problem, in which we would like to group samples into classes based on the similarity between the samples. We would like to create a neural network which will learn the class definitions for the known inputs, and will also be able to classify unknown inputs according to the learnt class definitions.

(a) Design an appropriate neural network model to complete the task.

(b) Discuss the advantages of the neural network model over classical k-means clustering algorithm for the above problem.