Learn Creating Interaction Features | Feature Engineering for Machine Learning

Definition

Interaction features are new variables formed by combining two or more existing features, often through mathematical operations such as multiplication, division, or addition, to reflect how these variables jointly influence the target.

Creating interaction features allows you to capture complex relationships between variables in the Titanic dataset, such as Age, Fare, Pclass, and Sex. The influence of one variable on survival can depend on another variable's value. For example, the effect of passenger class on survival may differ for males and females, or younger passengers might benefit more from higher fares. By combining features like Age * Fare or Pclass * Sex_encoded, you enable your model to learn these nuanced patterns, improving its ability to predict who survived based on how variables interact.


              1234567891011121314151617181920
            
import pandas as pd

# Sample Titanic-like dataset
data = {
    "Age": [22, 38, 26, 35, 28],
    "Fare": [7.25, 71.28, 7.92, 53.10, 8.05],
    "Pclass": [3, 1, 3, 1, 3],
    "Sex": ["male", "female", "female", "female", "male"],
    "Survived": [0, 1, 1, 1, 0]
}
df = pd.DataFrame(data)

# Encode 'Sex' as a numeric feature
df["Sex_encoded"] = df["Sex"].map({"male": 0, "female": 1})

# Create interaction features
df["Age_Fare_product"] = df["Age"] * df["Fare"]
df["Pclass_Sex_interaction"] = df["Pclass"] * df["Sex_encoded"]

print(df[["Age", "Fare", "Pclass", "Sex", "Age_Fare_product", "Pclass_Sex_interaction", "Survived"]])

Everything was clear?

Thanks for your feedback!

Section 3. Chapter 2

Ask AI

Ask anything or try one of the suggested questions to begin our chat

Suggested prompts:

Can you explain how these interaction features improve model performance?

What other interaction features could be useful for the Titanic dataset?

How do I interpret the values of these new features?

Awesome!

Completion rate improved to 8.33

Swipe to show menu

Definition


              1234567891011121314151617181920
            
import pandas as pd

# Sample Titanic-like dataset
data = {
    "Age": [22, 38, 26, 35, 28],
    "Fare": [7.25, 71.28, 7.92, 53.10, 8.05],
    "Pclass": [3, 1, 3, 1, 3],
    "Sex": ["male", "female", "female", "female", "male"],
    "Survived": [0, 1, 1, 1, 0]
}
df = pd.DataFrame(data)

# Encode 'Sex' as a numeric feature
df["Sex_encoded"] = df["Sex"].map({"male": 0, "female": 1})

# Create interaction features
df["Age_Fare_product"] = df["Age"] * df["Fare"]
df["Pclass_Sex_interaction"] = df["Pclass"] * df["Sex_encoded"]

print(df[["Age", "Fare", "Pclass", "Sex", "Age_Fare_product", "Pclass_Sex_interaction", "Survived"]])

Everything was clear?

Thanks for your feedback!

Section 3. Chapter 2