One-Hot Encoding
So, let's start to understand when and what encoding methods are best to use.
One-hot encoding is generally better to use when the categorical variable has no natural ordering or hierarchy between the categories and when the number of unique categories is relatively small. It is commonly used for nominal categorical data, where the categories have no inherent order or relationship between them.
Take a look at some examples of nominal categorical data:
- Colors: red, blue, green, yellow, etc.;
- Countries: USA, Canada, Mexico, Japan, etc.;
- Different pets: dog, cat, bird, fish, etc.;
- Genres of music: pop, rock, hip hop, country, etc.;
- Marital status: single, married, divorced, widowed, etc..
The basic idea behind one-hot encoding is to create a binary (0/1) variable for each category in the categorical variable.
We can perform one-hot encoding using the pd.get_dummies()
method, which creates 3 new binary columns for each of the three unique color values. The resulting dataset shows the binary representation of each color value:
Task
Use the one-hot encoding method on the 'cars.csv'
dataset.
Everything was clear?
Course Content
Data Preprocessing
2. Processing Quantitative Data
3. Processing Categorical Data
4. Time Series Data Processing
6. Moving on to Tasks
Data Preprocessing
One-Hot Encoding
So, let's start to understand when and what encoding methods are best to use.
One-hot encoding is generally better to use when the categorical variable has no natural ordering or hierarchy between the categories and when the number of unique categories is relatively small. It is commonly used for nominal categorical data, where the categories have no inherent order or relationship between them.
Take a look at some examples of nominal categorical data:
- Colors: red, blue, green, yellow, etc.;
- Countries: USA, Canada, Mexico, Japan, etc.;
- Different pets: dog, cat, bird, fish, etc.;
- Genres of music: pop, rock, hip hop, country, etc.;
- Marital status: single, married, divorced, widowed, etc..
The basic idea behind one-hot encoding is to create a binary (0/1) variable for each category in the categorical variable.
We can perform one-hot encoding using the pd.get_dummies()
method, which creates 3 new binary columns for each of the three unique color values. The resulting dataset shows the binary representation of each color value:
Task
Use the one-hot encoding method on the 'cars.csv'
dataset.
Everything was clear?