Date feature extraction in Data Analysis Python

Introduction

Date feature extraction helps us get useful parts from dates, like year or month, to understand patterns in data better.

You want to find sales trends by month or day.

You need to group data by weekdays or weekends.

You want to see if events happen more in certain years.

You want to add new columns like hour or quarter from timestamps.

Syntax

Data Analysis Python

df['new_feature'] = df['date_column'].dt.feature_name

Use .dt to access date parts from a pandas datetime column.

Common features: year, month, day, weekday, hour, quarter.

Examples

Extracts the year from the date column.

Data Analysis Python

df['year'] = df['date'].dt.year

Extracts the month number (1-12) from the date column.

Data Analysis Python

df['month'] = df['date'].dt.month

Extracts the weekday as a number (Monday=0, Sunday=6).

Data Analysis Python

df['weekday'] = df['date'].dt.weekday

Extracts the quarter of the year (1 to 4).

Data Analysis Python

df['quarter'] = df['date'].dt.quarter

Sample Program

This code creates a table with dates, converts them to date type, then adds new columns for year, month, weekday, and quarter.

Data Analysis Python

import pandas as pd

# Create sample data with dates
data = {'date': ['2023-01-15', '2023-04-20', '2023-07-10', '2023-10-05']}
df = pd.DataFrame(data)

# Convert 'date' column to datetime type
df['date'] = pd.to_datetime(df['date'])

# Extract year, month, weekday, and quarter
 df['year'] = df['date'].dt.year
 df['month'] = df['date'].dt.month
 df['weekday'] = df['date'].dt.weekday
 df['quarter'] = df['date'].dt.quarter

print(df)

Important Notes

Make sure your date column is in datetime format before extracting features.

Weekday numbers start at 0 for Monday, which helps in grouping by weekdays.

Quarter divides the year into four parts: Jan-Mar (1), Apr-Jun (2), Jul-Sep (3), Oct-Dec (4).

Summary

Date feature extraction breaks down dates into parts like year, month, and day.

It helps find patterns and group data by time periods.

Use pandas .dt accessor to get these features easily.