one class svm
时间: 2023-12-08 19:05:58 浏览: 28
One-class SVM is a type of support vector machine algorithm used for anomaly detection. It is a binary classification algorithm that separates data into two classes, normal and abnormal. However, in one-class SVM, the algorithm is trained only on the normal data and then it is used to classify new data as normal or abnormal. It is called "one-class" because it only needs one class of data to train, unlike traditional SVM which needs both positive and negative samples. One-class SVM finds a hyperplane that maximizes the margin between the normal data and the hyperplane. The distance between the hyperplane and the normal data is used to classify new data points. If the distance is within a certain threshold, the new data point is considered normal, otherwise it is considered abnormal.