Parametric Probabilistic Classification Using Matlab, write a function or script that implements a simple parametric classifier to produce plots similar to Figs 4.2 and 4.3 of Alpaydin. More specifically: Your code should take as input data for a one-dimensional, two-class classification problem, assuming a real-valued input feature. You should use Gaussian (normal) distributions as your model; i.e. fit a Gaussian to each class distribution using maximum likelihood estimates for the parameters of each Gaussian. This produces a model for p(x|C1) and for p(x|C2). Assume equal priors for the class distributions (P(C1) and P(C2)). In this case, the posteriors are just p(C1|x) and p(C2|x) and are equal to the discriminants g(x) for each class. To test your code, create a reduced version of the Iris dataset (see Prac 1 a plain text version of this dataset can be found in the practical materials); firstly by discarding all the features from the data except the first one (sepal length), secondly by discarding all examples for the third class (Iris-virginica). Using the Matlab hist() function, produce histograms to give you a picture of the data and something to compare you results with. Q4: Use your code to produce plots of the likelihoods and class posteriors for the reduced Iris data