One of the central steps in any data mining task to preprocess the data. This

includes normalization, handling missing values, feature selection, and so on$.$ So$,$

Alice wants to prepare the data first before she tries to apply any data mining

algorithm. To assist Alice in her quest, please answer the following questions:

$($a$)$ By looking into the data $($or your answer to $(2.$c$),$ do you think normalization

as described in class is important for this task? Why or why not? $($Hint:

do the attributes have the same scale? Shoud we treat a di

erence in GPA

from $3\mathrm{.}0$ to $4\mathrm{.}0$ the same as we treat a di

erence in age from $20$ to $21?)$

$($b$)$ The data contains some missing values. So$,$ Alice has to handle them some$-$

how. She considers the many options described in the class $($e$.$g$.$ eliminate,

replace, estimate, and so on$).$ By looking at the data, answer the following

questions:

Is it a good idea to eliminate the records with the missing values?

$($Hint: how will this impact the number of available records?$).$

Explain how Alice can estimate the missing values of the following

attributes: age, GPA, gender, pre$-$requisite, and pre$-$test score.