Customer Segmentation Model

This project demonstrates a Customer Segmentation system using K-Means Clustering. Customer segmentation helps businesses group their customers based on shared characteristics, allowing for tailored marketing strategies, improved customer satisfaction, and enhanced revenue opportunities.

---

Overview

The model groups customers into clusters based on their purchasing behaviors and other key features. Using K-Means Clustering, we determine an optimal number of clusters, which best fits the data, helping to identify natural groupings among customers.

---

Dataset Description

The dataset includes various customer features such as spending habits, income, and other demographic information. Each feature contributes to identifying segments with distinct purchasing or engagement patterns.

Key Features:

• CustomerID: Unique identifier for each customer
• Gender: Customer's gender
• Age: Age of the customer
• Annual Income: Estimated annual income of the customer
• Spending Score: Score assigned based on customer spending habits

---

Methodology

1. Feature Extraction: Extracted relevant features from the dataset needed for efficient clustering.
2. Elbow Method: Used the Elbow Method to find the optimal number of clusters. The elbow point on the plot indicates where the rate of decrease in inertia (within-cluster sum of squares) slows down, suggesting a good number of clusters.
3. K-Means Clustering: Applied K-Means with the optimal number of clusters found from the elbow method.

---

Model Usage

Below is an example of how to use this model to assign a new customer to a segment (this was not implemented in the actual code):

import numpy as np

# Example customer data: (Annual Income, Spending Score)
customer_data = (70000, 60)

# Convert customer data to numpy array
customer_np = np.asarray(customer_data)

# Reshape data for clustering (single instance)
customer_reshaped = customer_np.reshape(1, -1)

# Predict the cluster
cluster = model.predict(customer_reshaped)

print(f"The customer belongs to segment: {cluster[0]}")

Feel free to use or modify this model for customer segmentation tasks. Suggestions and contributions are welcome!