Clustering (K-Means)

# CLUSTERING: THE UNSUPERVISED APPROACH

import numpy as np
import matplotlib.pyplot as plt
from sklearn.datasets import make_blobs

# Generate sample data - like customer spending patterns
np.random.seed(42)
X, true_labels = make_blobs(
    n_samples=300, 
    centers=4,  # 4 hidden groups
    cluster_std=0.8,
    random_state=42
)

# Visualize the data WITHOUT labels
plt.figure(figsize=(12, 5))

plt.subplot(1, 2, 1)
plt.scatter(X[:, 0], X[:, 1], c='gray', alpha=0.6, edgecolor='black')
plt.title('Raw Data (No Labels)\nCan you see the clusters?')
plt.xlabel('Feature 1 (e.g., Annual Spending)')
plt.ylabel('Feature 2 (e.g., Frequency)')

# Show with true labels (what clustering tries to find)
plt.subplot(1, 2, 2)
plt.scatter(X[:, 0], X[:, 1], c=true_labels, cmap='viridis', 
            edgecolor='black', alpha=0.7)
plt.title('True Clusters (Hidden Structure)\nClustering reveals this!')
plt.xlabel('Feature 1')
plt.ylabel('Feature 2')

plt.tight_layout()
plt.show()

# Calculate cluster properties
print("\n" + "="*50)
print("CLUSTERING GOAL")
print("="*50)
print("✅ Minimize INTRA-cluster distance (tight groups)")
print("✅ Maximize INTER-cluster distance (well-separated)")