Clustering Algorithm for Customer Segmentation

A Step by Step Guide to K-Means Clustering

clustering analysis infographic (image by author from website)

What is Clustering Algorithm?

In a business context: Clustering algorithm is a technique that assists customer segmentation which is a process of classifying similar customers into the same segment. Clustering algorithm helps to better understand customers, in terms of both static demographics and dynamic behaviors. Customer with comparable characteristics often interact with the business similarly, thus business can benefit from this technique by creating tailored marketing strategy for each segment.

In a data science context: Clustering algorithm is an unsupervised machine learning algorithm that discovers groups of data points that are closely related. The fundamental difference between supervised and unsupervised algorithm is that:

• supervised algorithm: it requires partitioning the dataset into train and test set, and the algorithm learned based on the output/label of the train dataset and generalize it to unobserved data. For instance, decision tree, regression, neural networks.
• unsupervised algorithm: it is used to discover hidden patterns when there isn’t any defined output/label from the dataset. For instance, clustering, association rule mining, dimension reduction.