bad margin or suboptimal margin has
the hyperplane too close to one class or the distance not be as far as it can be to the
hyperplane for each point or support vector.
As for the one-class support vector machine, Figure
2-40
shows what the graph
would look like.
During training, the OC-SVM learns the decision boundary for normal observations,
accounting for a few outliers. If
novelties, new data points that the model has never seen
before, fall within this decision boundary, they are considered normal by the model. If
they fall outside of the boundary, they are considered anomalous. This technique is an
example of semi-supervised novelty detection, where the goal is to train the model on
normal data, and then it attempts to find anomalies in new data.
By doing so, the OC-SVM can capture the shape of the data pretty well thanks to the
decision boundary that captures most of the training observations.
Figure 2-40. An example of the decision boundary for a one-class support vector
machine
Chapter 2 traditional Methods of anoMaly deteCtion
63
Do'stlaringiz bilan baham: |