Kmean analysis

– Use the code, below, to run KMeans and to find out the best K for the given data set

– To find out the best K, use the KMeans inertia that is available using the kmeans.inertia_ instruction, below.

– Inertia is the mean squared distance between each instance and its closest centroid. This measures the “compactness” of all clusters. For more informaton on inertia you may refer to the reference on ML, Hands-on ML with Scikit learn, …

– Here is what needs to be done:

– 1. Run KMeans using K values ranging from 2 to 12, only, and record the inertia and cost/error in every run

– 2. Plot the recorded inertia values for K = 2 to 12

– 2. Plot the recorded cost (error) values for K = 2 to 12 

– 3. Do you see an “elbow” in either graph? if so, this is your best K. If not, decide on the “best” K and write why you picked this K value.

– 4. Plot the data with kmeans and the best K 

– 5. Open a new cell with a markdown cell type and write about your obeservations, how you picked the best K, and any conclusion.