Reference no: EM132757376

The table below lists a dataset containing details of policy holders at an insurance company. The descriptive features included in the table describe each policy holders' ID, occupation, gender, age, the type of insurance policy they hold, and their preferred contact channel. The preferred contact channel is the target feature in this domain.

ID OCCUPATION GENDER AGE POLICY TYPE PREF CHANNEL

1 lab tech female 43 planC email
2 farmhand female 57 planA phone
3 biophysicist male 21 planA email
4 sheriff female 47 planB phone
5 painter male 55 planC phone
6 manager male 19 planA email
7 geologist male 49 planC phone
8 messenger male 51 planB email
9 nurse female 18 planC phone

a. Using equal-frequency binning transform the AGE feature into a categorical feature with three levels: young, middle-aged, mature.

b. Examine the descriptive features in the dataset and list the features that you would exclude before you would use the dataset to build a predictive model. For each feature you decide to exclude explain why you have made this decision.

c. Calculate the probabilities required by a naive Bayes model to represent this domain.

d. What target level will a naive Bayes model predict for the following query: GENDER = female, AGE = 30, POLICY = planA