Which of the following models is useful for customer segmentation when labels are not available?

K-Means Clustering is an unsupervised machine learning algorithm that is particularly well-suited for customer segmentation when labeled data is not available. In this methodology, the model identifies patterns and groups in the data based solely on the features present, without any prior knowledge of labels.

When applied to customer segmentation, K-Means can analyze customer attributes such as purchasing behavior, demographics, or engagement metrics to group similar customers together. Each cluster that the algorithm forms represents a segment of customers that share common characteristics, which can then be targeted for tailored marketing strategies or specific product offerings.

The other models listed—Linear Regression, Random Forest, and Logistic Regression—are supervised learning techniques that rely on having labeled data for training. Linear Regression is used for predicting continuous outcomes; Random Forest is used for both classification and regression tasks but requires class labels; and Logistic Regression is specifically for binary classification tasks, also necessitating labeled data. These models are not ideal for the task of customer segmentation when no labels are available, thereby making K-Means the appropriate choice for this scenario.