customer_segmentation

Customer Segmentation using K-Means Clustering

Project Overview

This project applies K-Means Clustering to segment customers based on their shopping behavior. The dataset contains features such as age, gender, quantity purchased, price, payment method, and shopping mall. The goal is to identify distinct customer groups for better marketing strategies.

Dataset Information

• Columns:
  • Customer Id: Identification number of customer
  • Age: Age of the customer.
  • Gender: Male/Female.
  • Quantity: Number of items purchased.
  • Price: Total purchase amount.
  • Category: Category of items being purchased
  • Payment Method: How the customer paid (Card, Cash, etc.).
  • Shopping Mall: Mall where the purchase was made.

Methodology

1. Data Preprocessing

   • Handle missing values and data inconsistencies.
   • Convert categorical variables using One-Hot Encoding.
   • Standardize numerical features.

2. Exploratory Data Analysis (EDA)

   • Visualize distributions of age, quantity, and price.
   • Identify spending patterns by gender and mall.

3. Feature Engineering

   • Created new features like total spending.
   • Applied Principal Component Analysis (PCA) for dimensionality reduction.

4. K-Means Clustering

   • Determined optimal K value using Elbow Method & Silhouette Score.
   • Clustered customers into 3-5 segments.
   • Analyzed cluster characteristics.

5. Visualization

   • Scatter plots of customer segments.
   • Distribution of key attributes within clusters.

Key Insights

✔ Customer Segments Identified: [e.g., High Spenders, Budget Shoppers, Frequent Shoppers]
✔ Spending Trends: [e.g., Younger shoppers spend less but visit frequently.]
✔ Mall-Specific Preferences: [e.g., Mall A attracts high spenders, while Mall B has more budget-conscious buyers.]