Feature Engineering

What is Feature Engineering? #

Feature Engineering is the process of creating, selecting, and transforming variables (features) from raw data to improve model performance.

Why Feature Engineering is Important #

• Improves model accuracy
• Helps algorithms learn better patterns
• Reduces irrelevant data
• Converts raw data into useful features

Feature Selection #

🔹 What is Feature Selection? #

Selecting the most important features and removing unnecessary ones.

🔹 Why use it? #

• Reduces complexity
• Improves performance
• Saves computation time

🔹 Methods: #

• Filter Method: Select based on correlation
• Wrapper Method: Try different feature combinations
• Embedded Method: Feature selection during model training

🔹 Example: #

import pandas as pd

data = pd.read_csv("data.csv")

# Select only important columns
selected_features = data[["age", "salary"]]
print(selected_features.head())

Feature Extraction #

🔹 What is Feature Extraction? #

Creating new features from existing data.

🔹 Examples: #

• Extract year from date
• Create total price = quantity × price
• Convert text into numbers

🔹 Python Example: #

data["total_price"] = data["quantity"] * data["price"]

# Extract year
data["year"] = pd.to_datetime(data["date"]).dt.yeardata["total_price"] = data["quantity"] * data["price"]

# Extract year
data["year"] = pd.to_datetime(data["date"]).dt.year

Encoding (Categorical Data) #

🔹 What is Encoding? #

Converting categorical (text) data into numerical form.

🔹 Why needed? #

Machine learning models cannot understand text directly

Types of Encoding #

Label Encoding #

from sklearn.preprocessing import LabelEncoder

le = LabelEncoder()
data["city"] = le.fit_transform(data["city"])

One-Hot Encoding

data = pd.get_dummies(data, columns=["city"])