Centroid Table for Q1

Attributes	Cluster_0	Cluster_1	Cluster_2	Cluster_3	Cluster_4
Balance	0.546	1.846	0.250	1.391	-0.470
Balance_Frequency	0.434	0.306	-0.040	0.434	-0.096
Purchases	1.327	10.877	-0.366	0.001	-0.102
One-off Purchases	1.131	10.177	-0.247	-0.220	-0.164
Installment Purchases	1.059	7.011	-0.412	0.407	0.060
Cash Advance	-0.016	0.379	0.363	0.243	-0.397
Purchase Frequency	1.120	1.040	-0.907	0.042	0.636
One-off Purchases Frequency	1.622	1.863	-0.417	-0.451	-0.054
Purchases Installments Frequency	0.901	0.989	-0.793	0.242	0.577
Cash_Advance_Frequency	-0.133	-0.282	0.526	-0.162	-0.529
Cash_Advance_Trx	-0.050	0.034	0.379	0.060	-0.398
Purchases_Trx	1.626	5.197	-0.498	0.219	-0.002
Credit_Limit	0.983	3.010	-0.046	0.076	-0.278
Payments	0.806	7.947	-0.036	0.030	-0.276
Minimum_Payments	0.117	1.059	0.001	11.358	-0.164
Prc_Full_Payment	0.294	1.052	-0.426	-0.538	0.370
Tenure	0.284	0.306	-0.094	0.293	0.010

K Means	Overall Avg within Centroid Distance	Variance from K to K
5	-11.139	-1.412
6	-9.727	-0.096
7	-9.631	-0.978
8	-8.653	-0.407
9	-8.246	-0.434
10	-7.812	N/A

Cluster Model for Q2b:

Cluster	Items
Cluster 0:	2757 items
Cluster 1:	125 items
Cluster 2:	24 items
Cluster 3:	604 items
Cluster 4:	1227 items
Cluster 5:	656 items
Cluster 6:	36 items
Cluster 7:	3207 items
Total number of items:	8636

1-Which k value is appropriate for this problem and why? (Again, offer analysis of what you think of an appropriate k value based on your experiments with different k values and the plot using the elbow method.)

2- reassign meaningful names to the clusters (instead of generic names such as cluster_0, cluster_1, etc.) based on the characteristics of the segmentation (Offer your own analysis of what you think of clusters based on those numbers)