Unsupervised statistical learning (USL) techniques, such as self-organizing maps (SOMs), principal component analysis (PCA) and independent component analysis explore di®erent statistical properties to e±ciently process information from multiple variables. USL algorithms have been successfully applied in experimental high-energy physics (HEP) and related areas for di®erent purposes, such as feature extraction, signal detection, noise reduction, signal-background separation and removal of cross-interference from multiple signal sources in multisensor measurement systems. This paper presents both a review of the theoretical aspects of these signal processing methods and examples of some successful applications in HEP and related areas experiments.