Description
Extra Credit (for an additional 20 points): – Use your prior
implementation of k-means to find prototypes to act as proxy examples
in k-nearest neighbor. Note that the k for k-means need not be the
same as the k for k-nearest neighbor. Be sure to tune to find the
proper number of clusters. You will use this on both the
classification and the regression problems. – Implement a radial basis
function neural network using one hidden node for each data point from
a random sample of 10% of the training set. – Implement the radial
basis function neural network using the results of k-means clustering
for the hidden nodes. You may use the results of clustering for
prototypes here.
