Machine Learning $-$ K$-$Means Clustering

Objective:

To design and evaluate the performance of a k$-$means clustering model on a synthetic dataset.

Instructions:

$1.$ Generate the synthetic dataset as follows:

make$\_$blobs$($n$\_$samples$=600,$ centers$=4,$ n$\_$features$=2,$ random$\_$state$=20)$

$2.$ Find the optimal number of centroids according to the Elbow method. Justify your

selection.

$3.$ Find the optimal number of centroids according to the Silhouette method. Justify your

selection. Which method is more accurate?

$4.$ Fit a k$-$Means model, with k$=4.$ Visualize all clusters found by k$-$Means. Report the

inertia of the model. Analyze and discuss your plot.

$5.$ Evaluate the accuracy of labels estimated by k$-$Means $($Hint: you may use the

classification$\_$report$()$ method$).$ Analyze the classification report and draw your

conclusions for the designed model.

Note$-$ Include the code and explanation