On-line unsupervised outlier detection using finite mixtures with discounting learning algorithms.
Kenji YamanishiJun'ichi TakeuchiGraham J. WilliamsPeter MilnePublished in: KDD (2000)
Keyphrases
- outlier detection
- finite mixtures
- minimum message length
- change point detection
- learning algorithm
- density estimation
- finite mixture model
- unsupervised learning
- supervised learning
- mixture model
- knowledge discovery
- detection algorithm
- machine learning algorithms
- expectation maximization
- detecting outliers
- data streams
- density ratio estimation
- machine learning
- em algorithm
- training data
- gaussian mixture model
- learning problems
- active learning
- semi supervised learning
- semi supervised
- data mining
- training examples
- unlabeled data
- information retrieval