Creating interaction features in Data Analysis Python - Visual Walkthrough

Concept Flow - Creating interaction features

Start with dataset

↓

Select two or more features

↓

Multiply or combine features

↓

Create new interaction feature

↓

Add new feature to dataset

↓

Use enhanced dataset for modeling

We start with data, pick features to combine, multiply or combine them to create new features, then add these to the dataset for better analysis.

Execution Sample

Data Analysis Python

import pandas as pd

df = pd.DataFrame({'A': [1, 2], 'B': [3, 4]})
df['A_B'] = df['A'] * df['B']
print(df)

This code creates a new feature by multiplying columns A and B, then adds it to the dataframe.

Execution Table

Step	Action	DataFrame State	New Feature 'A_B'
1	Create DataFrame with columns A and B	{'A': [1, 2], 'B': [3, 4]}	Not created
2	Multiply A and B to create 'A_B'	{'A': [1, 2], 'B': [3, 4], 'A_B': [3, 8]}	3, 8
3	Print DataFrame	{'A': [1, 2], 'B': [3, 4], 'A_B': [3, 8]}	3, 8
4	End of execution	Final DataFrame with interaction feature	3, 8

💡 All steps completed, interaction feature 'A_B' added successfully.

Variable Tracker

Variable	Start	After Step 1	After Step 2	Final
df	None	{'A': [1, 2], 'B': [3, 4]}	{'A': [1, 2], 'B': [3, 4], 'A_B': [3, 8]}	{'A': [1, 2], 'B': [3, 4], 'A_B': [3, 8]}

Key Moments - 2 Insights

Why do we multiply columns A and B to create the new feature?

Multiplying columns A and B captures the combined effect of both features, which might reveal patterns not visible when using them separately. See execution_table step 2 where 'A_B' is created.

Is the original data changed when we add the interaction feature?

No, the original columns A and B remain unchanged. The new feature 'A_B' is added as an extra column, as shown in execution_table step 2.

Visual Quiz - 3 Questions

Look at the execution_table at step 2, what is the value of 'A_B' for the second row?

A 8

B 3

C 6

D 4

💡 Hint

Check the 'New Feature A_B' column in execution_table row for step 2.

At which step is the new interaction feature added to the DataFrame?

A Step 3

B Step 1

C Step 2

D Step 4

💡 Hint

Look at the 'Action' column in execution_table to see when 'A_B' is created.

If we changed the operation from multiplication to addition, how would the 'A_B' values change?

A They would be the difference of A and B values

B They would be the sum of A and B values

C They would be the product of A and B values

D They would remain the same

💡 Hint

Think about what addition does compared to multiplication on the columns.

Concept Snapshot

Creating interaction features:
- Combine two or more features (e.g., multiply)
- Add the result as a new column
- Helps capture combined effects
- Use: df['new'] = df['A'] * df['B']
- Enhances model input data

Full Transcript

Creating interaction features means combining existing data columns to make new ones that show combined effects. For example, multiplying two columns creates a new feature that can help models find patterns better. We start with a dataset, pick columns, multiply them, and add the result as a new column. This new column does not replace the old ones but adds more information. The example code creates a new column 'A_B' by multiplying columns 'A' and 'B'. The execution table shows each step, including creating the DataFrame, adding the new feature, and printing the result. Variables like the DataFrame change only when the new feature is added. Common confusions include why multiply and whether original data changes. The quizzes check understanding of values and steps.