Non-derivable itemsets for fast outlier detection in large high-dimensional categorical data.
Anna KoufakouJimmy SecretanMichael GeorgiopoulosPublished in: Knowl. Inf. Syst. (2011)
Keyphrases
- categorical data
- outlier detection
- high dimensional
- high dimensional datasets
- distance based outlier detection
- frequent itemset mining
- itemsets
- frequent itemsets
- parameter free
- cluster analysis
- data streams
- numerical data
- condensed representations
- binary data
- fraud detection
- low dimensional
- detection algorithm
- knowledge discovery
- data mining
- dimensionality reduction
- detecting outliers
- density based clustering
- similarity search
- continuous data
- multi dimensional
- attribute values
- high dimensionality
- pattern mining
- density ratio estimation
- feature space
- unsupervised learning
- nearest neighbor
- detect outliers
- databases
- outlier mining
- pairwise
- high dimensional spaces
- association rule mining
- high dimensional data
- association rules