4. a) The following is a scatter plot of data collated by scientists based on age and salary levels of 11 individuals. A linear classifier was created to predict high and low income earners based on this data set in the form of z = a + b1* (X1) + b2 * (X2) where X1 = Income level in $'000 and X2 = Age of interviewees. Derive the boundary line equation. 60 Age (X2) 50 40 30 20 10 0 0 5 10 A B 15 D E 20 25 30 Income Level $'000 (X1) (10 marks) b) An analyst downloaded the current revenues achieved vs. the amount of $ investment spent in advertisement and came up with the following statistics. X Amount of $ spent on advertisement ($'000) 5 15 10 20 20 Revenue ($'000) 20 35 20 35 10 20 40 45 The analyst wanted to apply K-means clustering on the data to get a clear idea of these shoppers' behavior. Apply a K-2 cluster, with starting centroids of AB and CE and 1 iteration, calculate the mean of the ending centroids and clusters created. 50