This notebook demonstrates a K-Nearest Neighbors classification workflow on a telecommunications customer dataset, segmented into four service-usage groups. It uses pandas for data loading and preprocessing, and scikit-learn’s `preprocessing` module for feature scaling. The core visualization combines `matplotlib.pyplot` scatter plots to display demographic features (such as region and age) colored by customer category, alongside a manually inserted diagram of the K-NN algorithm. The analysis shows how varying the value of K affects class predictions for unknown data points.