One-Class SVM

One-Class SVM is a machine learning method used to identify whether data points belong to a specific "normal" category or are different (anomalies). It works by learning the typical pattern of normal data and then checking new data against this pattern. If the new data fits well, it's considered normal; if not, it's flagged as an anomaly. This approach is useful in security, fraud detection, and quality control, especially when only normal data are available for training, allowing the system to spot unusual or rare instances effectively.