$($b$)$ In this part, we do an experiment to explore the use of clustering for making predictions. $[11]$

$($i$)$ Write code to apply k$-$means to cluster the scikit$-$learn$$s breast cancer dataset $($see the load$\_$breast$\_$cancer$()$ function$)$ to $2$ clusters, using only the features but not the target. Also write code to$,$ for each of the resulting clusters, count the data points $($i$.$e$.$ instances$)$ with target value $0,$ and those with target value $1$; display the counts.

$($ii$)$ Notice that if the clusters have different majority target classes, the clustering scheme may be used for predicting a data point as the majority target class of the cluster that data point belongs to$.$ Compute the accuracy using the total count of the majority target classes and the total number of data points.

$($c$)$ Repeat part $($b$)$ using agglomerative hierarchical clustering instead of k$-$means clustering.