Encoding categorical data

In some data set there may be categorical data. For example we can consider following data set.

Country     Age     Salary
USA            35        $62,000
UK              25        $58,000
Canada        40        $69,000
USA            40        $78,000
UK              30        $64,000
Canada        30        $61,000

If we consider the country column, it can be categorized by country. We are going to categorize it by giving USA=1, UK=2 and Canada=3.
* Note: Since we are going to build mathematical model, we can't deal with data like row values in Country column. Therefore we convert them to numerical values.

Country     Age     Salary
1                  35        $62,000
2                  25        $58,000
3                  40        $69,000
1                  40        $78,000
2                  30        $64,000
3                  30        $61,000

Now we have a categorized data set. But still we have a problem. The problem is while USA=1 Canada takes 3 then if we put these values in a mathematical calculation, Canada will take higher value than USA although it should not be like that(according to this eg). How to prevent this in ML? The solution is Dummy Encoding or dummy variables.

Dummy Encoding
In dummy variables concept, the categorical column is replaced with new columns corresponding to the number of categories. For eg: The Country column will be replaced with USA,UK and Canada columns.

USA     UK   Canada   Age     Salary
1             0          0          35        $62,000
0             1          0          25        $58,000
0             0          1          40        $69,000
1             0          0          40        $78,000
0             1          0          30        $64,000
0             0          1          30        $61,000


Encoding in Python
# Encoding categorical data
from sklearn.preprocessing import LabelEncoder, OneHotEncoder
labelencoder_X = LabelEncoder()
X[:, 0] = labelencoder_X.fit_transform(X[:, 0])
onehotencoder = OneHotEncoder(categorical_features = [0])
X = onehotencoder.fit_transform(X).toarray()

Encoding in R
# Encoding categorical data
dataset$Country = factor(dataset$Country,
                         levels = c("USA", "UK", "Canada"),
                         labels = c(1,2,3))



Comments

Popular posts from this blog

Importing libraries in Python

Taking care of missing data in a data set