Given a data sample of key stroke latency (in milliseconds) K = {100, 400, 200, 350, 200, 3500, 500, 650, 800, 900, 400, 200, 2500, 100, 100, 200, 300, 200, 300, 550}

(a) (6 points) Using mean and standard deviation of the data sample to detect the outlier. (Assume that the weight for standard deviation = 2) (Hint: In this method, an observation is considered as an outlier when it is not with in the range of mean (weight * standard deviation))

(b) (6 points) Using the Distance-based outlier detection method to detect the outlier, where = 0.8 and r = 150.

(Hint: In this method, an observation is considered as an outlier when at least a fraction of the observations are further than r from it. Use the definition d(i,j) = |xi xj | to represent the distance between two points xi and xj )