Clustering Mixed Data with K-Means: From FAMD to Segments

Applying K-Means to FAMD-transformed data, selecting the optimal number of clusters and evaluating the final segmentation. Photo by Jeyakumaran Mayooresan on Unsplash Introduction In the second part of the series, we transformed our dataset using Factor Analysis of Mixed Data (FAMD) to create a meaningful lower-dimensional representation of mixed variables. In this article, we take the next step and apply K-Means clustering to this transformed dataset in order to identify distinct groups of customers . We will focus on the technical aspects of clustering, including how to determine the optimal number of clusters and how to fit the model . The resulting clusters will form the foundation for the insights and business interpretations that will be explored later in this project. Finally, don’t forget to check out the previous articles in this series for a complete view of the end-to-end project: Part 1: From PCA to FAMD: Dimensionality Reduction for Mixed Data Part 2: Applying FAMD in Pyth