Question $13(1$ point$)$

Suppose you have been hired as a data scientist by a Technical Help$-$Line company.

People can call this company for help with any computer$-$related issues or questions.

The service is free for the first call and then $$\frac{5}{?}$ minute for any follow$-$up calls. As a

first step, the company wants to try to predict which callers will call again $($so will be

follow$-$up and therefore be paying customers$).$ For each first$-$time caller, the

company collects data on the person's age, call duration, number of questions asked

by caller during the call, years of employment, and yearly income. You also have data

on which first$-$time callers call again within a $6-$month window of time.

Choose the statement that is least likely, is not true, or is least helpful for the

company and your goals.

a$)$ The labeled dataset you are building will have five independent numeric

variables that can be called AGE, CALL$\_$Minutes, NUM$\_$QUEST, YRS$\_$EMP,

YR$\_$INCOME.

b$)$ Your goal is the use the data for the $5$ variables, along with the label to predict

whether a caller will call again during the $6-$month time frame. Therefore, the

rows of your dataset are the callers, the label $($dependent or target variable$)$ is

Repeat or Single, and you have $5$ columns of data $($numeric$).$

c$)$ All $6$ variables are likely to be equally correlated with whether a caller will be a

repeat caller and so there is no need to investigate each variable so as to

reduce the dimensionality of the dataset.

d$)$ As a first step, you build a dataset to represent first$-$time callers. This dataset

will be labeled. The label $($also thought of as the target or dependent variable$)$

will have two options, Repeat or Single. The "Repeat" implies that the first$-$

time caller called again $($and so became a paying customer$)$ within the $6-$

month time frame. The "Single" implies that the caller never called again

within the $6-$month time frame.

e$)$ Because your label or target to predict is "Single" or "Repeat", you determine

that you can try to model the data with logistic regression.