Telecommunications companies providing cell-phone service are interested in customer retention. In particular, identifying customers who are about to churn (cancel their service) is potentially worth millions of dollars if the company can proactively address the reason that customer is considering cancellation and retain the customer. The Cellphone data contains 1000 customer data to be used to classify a customer as a churner or not.

Partition the data with all the variables with Oversampling procedure, so that there is 50% successes (churners) in the training set and 40% of the validation data are taken away as test set. Use 12345 as the seed in the randomized sampling.

Classify the data using k-Nearest Neighbors with up to k = 15. Use Churn as the output variable and all the other variables as input variables. First normalize Input Data.

a) What is the overall error rate, the Class 1 and Class 0 error rates, the sensitivity and specificity for the test set?

b) How many false positives and false negatives did the model commit on the test set? What percentage of predicted churners were false positives? What percentage of predicted nonchurners were false negatives?

c) Classify the following ten customers as churners or not.

Account Weeks	Contract Renewal	Data Plan	Data Usage	CustServ Calls	Day Mins	Day Calls	Monthly Charge	Overage Fee	Roam Mins
100	0	0	0	1	12,5	70	24	12,83	7,7
126	1	1	0	2	193	97	41	4,49	15,3
130	1	1	2,75	1	174,5	103	80,1	12,28	10,3
129	0	1	3,73	4	292,9	101	104,3	9,99	12,8
160	0	0	0	1	152,7	105	48	13,22	9,4
70	0	1	3,22	1	106,7	78	68,7	10,49	12,1
124	1	0	0	2	243,8	98	49	4,2	13,7
123	1	0	0	5	159,1	114	47	11,57	8,8
123	1	1	1,57	3	198,4	131	56,7	3,77	5,8
140	1	0	0	4	178,9	101	45	8,46	13,8