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 50 40 30 Age (X2) 20 10 0 0 0 5 10 15 35 40 45 50 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 Y Amount of $ spent on advertisement Revenue ($'000) ($'000) A 5 20 B 15 35 10 20 D 20 10 E 20 20 00 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. 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 50 40 30 Age (X2) 20 10 0 0 0 5 10 15 35 40 45 50 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 Y Amount of $ spent on advertisement Revenue ($'000) ($'000) A 5 20 B 15 35 10 20 D 20 10 E 20 20 00 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