Outlier Detection for Large High-Dimensional Categorical Data using Non-Derivable and Non-Almost-Derivable Sets.
Anna KoufakouJimmy SecretanMichelle FoxGary GramajoGeorgios C. AnagnostopoulosMichael GeorgiopoulosPublished in: DMIN (2009)
Keyphrases
- outlier detection
- categorical data
- distance based outlier detection
- parameter free
- high dimensional
- high dimensional datasets
- cluster analysis
- density based clustering
- numerical data
- detection algorithm
- categorical attributes
- detecting outliers
- fraud detection
- density ratio estimation
- density estimation
- dimensionality reduction
- data streams
- knowledge discovery
- data mining
- database
- nearest neighbor
- attribute values
- low dimensional
- credit card fraud detection
- multi dimensional
- detect outliers
- high dimensionality
- clustering algorithm
- similarity search