user@devops:~$ cat README.md
K-Means — Customer Segmentation
# Description
Customer segmentation project with K-Means Clustering, introducing UNSUPERVISED learning. Simulates 500 mall customers with 4 realistic profiles (youth, working adults, seniors, students). Learn the K-Means algorithm step by step, the elbow method and silhouette score for choosing K, the critical importance of scaling with StandardScaler, customer profile interpretation per cluster, and visualization with PCA. Includes inference for automatically classifying new customers.
# Key features
$ K-Means algorithm: centroids, assignment, iteration until convergence
$ Elbow Method for visual K selection
$ Silhouette Score: quantitative metric for cluster separation
$ Scaling with StandardScaler: critical for K-Means
$ Customer profile interpretation per cluster
$ 2D visualization with PCA and cluster map
$ Inference: automatic classification of new customers
# Gallery
# Technologies used
