Problem #4115 points!: kNN Suppose you want to build a nearest neighbors classifier to predict whether a patient who had undergone surgery for breast cancer can survive 5 years or longer. You collect the following data: 50 50 50 51 51 51 52 52 52 53 53 53 54 54 54 56 56 56 57 57 58 59 Age #Positive 13 4 1 3 1 3 4 0 12 1 0 5 7! 0: 31 0 0 14 0 335 Survival2 1 1 2 1 1 2 1 1 2 1 12 1 1 2 1 1 212 (Age: age of patient at time of operation; #Positive: number of positive axillary nodes detected; Survival:-the patient survived 5 years or longer, 2 - the patient died within 5 years) (a) [5 pts] Based on the kNN method, given k-1,2, 5, what is the label of a test point with Age 55 and #Positive-3? Are the results the same given different k? Why? (b) [5 pts] What do you think is the best value for k in this case? Why? (c) [5 pts] Is the point (Age-59, #Positive-35) in the training set an outlier? Why? If it is, how would you detect it? Problem #5 15 points!: Clustering en the k-means and the k-medoids algorithms for clustering. Problem #4115 points!: kNN Suppose you want to build a nearest neighbors classifier to predict whether a patient who had undergone surgery for breast cancer can survive 5 years or longer. You collect the following data: 50 50 50 51 51 51 52 52 52 53 53 53 54 54 54 56 56 56 57 57 58 59 Age #Positive 13 4 1 3 1 3 4 0 12 1 0 5 7! 0: 31 0 0 14 0 335 Survival2 1 1 2 1 1 2 1 1 2 1 12 1 1 2 1 1 212 (Age: age of patient at time of operation; #Positive: number of positive axillary nodes detected; Survival:-the patient survived 5 years or longer, 2 - the patient died within 5 years) (a) [5 pts] Based on the kNN method, given k-1,2, 5, what is the label of a test point with Age 55 and #Positive-3? Are the results the same given different k? Why? (b) [5 pts] What do you think is the best value for k in this case? Why? (c) [5 pts] Is the point (Age-59, #Positive-35) in the training set an outlier? Why? If it is, how would you detect it? Problem #5 15 points!: Clustering en the k-means and the k-medoids algorithms for clustering