Question

Given the min-max normalization formula is given as follows: New value = [(value - min)/(max-min)] * (new_max - new_min) + new_min. 1. Consider we have values of age attribute in the range of 20-80. Normalise an age value 30 using the min-max normalization into [10,20). 2. The ages of 5 individuals are (20,30,25,50,25). Normalise the average of these individuals using the min-max normalisation into [0,1].

Answer 1

Solution: Min Max Normalization is one of the data preprocessing methods used in data mining for rescale data within small range of required:

Formula is shown below:

new value  = [(value - min)/(max - min)] * (new max-new min) + new min

Answer to Q1) Given Range of age values :20-80 ie min=20, max = 80, find min max normalization value for age 30 with new range[10,20] ie new min=10, new max =20

Substitute these values in the formula:

new value  = [(value - min)/(max - min)] * (new max-new min) + new min

=[(30-20)/(80-20)] * (20-10)+10

= [10/60]* 10+10

= (1/6) * 20 =3.33

Ans is 3.33

Answer to Q2)

Ages of 5 people is given as (20,30,25,50,25)

Average age is: 20 + 30 + 25 + 50 + 25 =150/5=30

So for average age value 30, the min max normalised value for range [0.1] is shown below:

value = 30

min=20, max=50, new min = 0, new max =1

by substitute in the min max formula:

new value  = [(value - min)/(max - min)] * (new max-new min) + new min

= [(30-20)/(50-20)] * (1-0)+0

= [(10/30)] * (1)

= 0.33