StackedML
Practice
Labs
Questions
Models
Pricing
Sign in
Questions
/
Unsupervised Learning
/
Other Concepts
/
Anomaly detection
← Previous
Next →
553.
One-Class SVM for Anomaly Detection
medium
A one-class SVM is trained to detect anomalies. What does it learn?
A
A probability distribution over the normal data and flags points with low likelihood as anomalies
B
A boundary around the normal data in feature space — points outside the boundary are flagged as anomalies
C
A distance metric that measures how far each point is from the centroid of the normal class
D
A decision hyperplane between normal and anomalous classes using only normal training examples during the training process
Sign in to verify your answer
← Back to Questions