An online retailer combined many data sources into a single table with 980 variables. You need to build a model to discover groups of customers likely to respond to a marketing campaign. Which pair of techniques should you use?

K-means clustering is appropriate for discovering groups of customers (customer segmentation) based on their characteristics, which aligns with the goal of finding groups likely to respond to a marketing campaign. Principal Component Analysis (PCA) is a dimensionality reduction technique essential for handling the large number of variables (980) by transforming them into a smaller set of uncorrelated components while retaining most of the variance. This reduces computational complexity and mitigates the curse of dimensionality, making K-means more effective. Latent Dirichlet Allocation (LDA) is primarily used for topic modeling in text data, not for customer segmentation on tabular data. Semantic segmentation is a computer vision technique for pixel-level classification in images. Factorization machines (FM) are used for recommendation systems and predicting interactions between features, not directly for unsupervised customer grouping.