An unsupervised method that groups data into clusters based on similarity. In forex, clustering algorithms (such as K-means or hierarchical clustering) can segment historical market conditions into regimes (e.g. trending vs. range-bound). Clustering assigns each data point (such as a multi-indicator snapshot) to a cluster so that similar points share a cluster. A trading system might use clustering to identify recurring patterns; for instance, grouping time periods by volatility. Clustering is often the first step in feature engineering or anomaly detection, highlighting common vs. anomalous behavior in price data.

Back to Robots Glossary Index