Customer Segmentation - Pattern recognition using k means clustering algorithm

The recent project, PATTERN RECOGNITION IN RETAIL DATA, utilized advanced techniques such as the K-means clustering algorithm alongside the elbow method and silhouette score to determine the optimal number of clusters for the dataset. The project aims to showcase how these techniques can be used to identify patterns and insights within retail data, ultimately aiding businesses in making informed decisions. Here customers have been segmented to 3 clusters based on country's revenue generation, potentially leading to high, medium and low revenues.

A catch, the customer segmentation was done but there's an issue in visualizing the clusters/segments due to a data outlier. The country is a categorical column against which revenues a numerical column is used for plotting and that has caused a overfitted or incorrect plot to visual our customer clusters, which needs to be corrected, as part of next phase. Stay Tuned