Answer Happy

3. (Total: 6 points) Recall our discussion of machine learning application in class, and the specific problem of evaluation. We have customer information based on age and income, and normalize the information. Age Income Customer John Rachel Hannah Bentancur Tom Nellie Kane David 0.35 0.34 1 0.81 0.93 0.39 0.27 0.78 0.35 0.25 0.70 0.48 1 0.2 0.55 0.25 In order to group our customers into 2 clusters, we want to randomly select the two initial centroids and let's say those are: initial centroid 1: John and initial centroid 2 David. Our goal is to find two centroid and customers who belong to each of the centroids after 2 iterations. Please fill out these contingency tables.
x diff^2 y diff^2 Distance 1" iteration: Centroid 1: John Age Income Location Income Customer Age difference difference John Rachel Hannah Bentancur Tom Nellie Kane David Centroid 2: David Age Income Location Income Customer Age difference difference John Rachel Hannah Bentancur Tom Nellie Kane David x diff^2 y diff^2 Distance Distance from 1 Distance from 2 Customer John Rachel Hannah Bentancur Tom Nellie Kane David
Age Income Customer John Rachel Hannah Bentancur Tom Nellie Kane David New Centroid 1 Age Income Customer John Rachel Hannah Bentancur Tom Nellie Kane David New Centroid 2
2nd iteration: Centroid 1: John Age Location Income Income difference Age difference x diff^2 y diff ^2 Distance Customer John Rachel Hannah Bentancur x diff A2 y diff A2 Distance Tom Nellie Kane David Centroid 2: David Age Income Location Income Customer Age difference difference John Rachel Hannah Bentancur Tom Nellie Kane David Distance from 1 Distance from 2 Customer John Rachel Hannah Bentancur Tom Nellie Kane David
Age Income Customer John Rachel Hannah Bentancur Tom Nellie Kane David New Centroid 1 Age Income Customer John Rachel Hannah Bentancur Tom Nellie Kane David New Centroid 2 Conclusion: Centroid 1 Location: Customers in Centroid 1: Centroid 2 Location: Customers in Centroid 2: