MCMD2 : Command Reference : mnormalize - Normalization

3.35 mnormalize - Normalization

Specify the field at the f= parameter, and specify the normalization method at c= parameter.

Format

mnormalize c= f= [k=] [i=] [o=] [bufcount=] [-nfn] [-nfno] [-x] [-q] [precision=] [--help] [--version]

Parameters

`c=`	Specify the normalisation method listed as follows.
	`z` : z score : $z_ i=(x_ i-m)/u$ ( $x_ i$ : $i$ number of data, $m$ :arithmetic mean, $u$ :standard deviation)
	`Z` : deviation value : $Z_ i=50+10\times z_ i$
	`range` : use linear conversion to transform minimum value 0 to maximum value 1 $r_ i=(x_ i-\min _ x)/(\max _ x-\min _ x)$
`f=`	Specify the field to normalize here.
	Specify the new field name after :(colon). Example: `f=`quantity:quantityNorm
`k=`	Key field name(s) [aggregate key break processing]
	The key field specified is used as the unit for normalization.

Examples

Example 1: Basic Example

Normalize (z score) quantity and amount field based on each customer, label the column names of the output as qtyNominal and amtNorminal respectively.

$ more dat1.csv
customer,quantity,amount
A,1,10
A,2,20
B,1,15
B,3,10
B,1,20
$ mnormalize c=z k=customer f=quantity:qtyNominal,amount:amtNorminal i=dat1.csv o=rsl1.csv
#END# kgnormalize c=z f=quantity:qtyNominal,amount:amtNorminal i=dat1.csv k=customer o=rsl1.csv
$ more rsl1.csv
customer%0,quantity,amount,qtyNominal,amtNorminal
A,1,10,-0.7071067812,-0.7071067812
A,2,20,0.7071067812,0.7071067812
B,1,15,-0.5773502692,0
B,3,10,1.154700538,-1
B,1,20,-0.5773502692,1

Example 2: Deviation value

$ mnormalize c=Z k=customer f=quantity:qtyNominal,amount:amtNorminal i=dat1.csv o=rsl2.csv
#END# kgnormalize c=Z f=quantity:qtyNominal,amount:amtNorminal i=dat1.csv k=customer o=rsl2.csv
$ more rsl2.csv
customer%0,quantity,amount,qtyNominal,amtNorminal
A,1,10,42.92893219,42.92893219
A,2,20,57.07106781,57.07106781
B,1,15,44.22649731,50
B,3,10,61.54700538,40
B,1,20,44.22649731,60

Example 3: Linear transformation from 0 to 1

$ mnormalize c=range k=customer f=quantity:qtyNominal,amount:amtNorminal i=dat1.csv o=rsl3.csv
#END# kgnormalize c=range f=quantity:qtyNominal,amount:amtNorminal i=dat1.csv k=customer o=rsl3.csv
$ more rsl3.csv
customer%0,quantity,amount,qtyNominal,amtNorminal
A,1,10,0,0
A,2,20,1,1
B,1,15,0,0.5
B,3,10,1,0
B,1,20,0,1

Related Command