OAB - An Open Anomaly Benchmark Framework for Unsupervised and Semisupervised Anomaly Detection on Image and Tabular Data Sets.
Andreas LohrerJan DellerMaximilian HünemörderPeer KrögerPublished in: ICDM (Workshops) (2021)
Keyphrases
- anomaly detection
- unsupervised anomaly detection
- data sets
- anomalous behavior
- intrusion detection
- detecting anomalies
- input image
- unsupervised learning
- semi supervised
- network intrusion detection
- network anomaly detection
- similarity measure
- network traffic
- intrusion detection system
- behavior analysis
- network security
- machine learning
- supervised learning
- network intrusion
- real world
- connectionist systems
- data assimilation
- detecting anomalous
- computer security
- image segmentation
- detect anomalies
- computer vision
- one class support vector machines
- feature extraction
- negative selection algorithm
- training data
- pattern recognition
- object recognition
- training set