A Method for Modeling Data Anomalies in Practice.
Jennifer HorkoffMiroslaw StaronWilhelm MedingPublished in: SEAA (2021)
Keyphrases
- synthetic data
- pairwise
- input data
- missing data
- test data
- data collection
- prior knowledge
- statistical methods
- database
- significant improvement
- cost function
- raw data
- data sets
- high precision
- high accuracy
- prior information
- spectral clustering
- computational cost
- neural network
- noisy data
- user input
- modeling method
- correlation analysis
- data distribution
- rare events
- detection method
- unsupervised learning
- model selection
- knowledge discovery
- computational complexity
- training data
- high dimensional data
- labeled data
- data mining techniques
- support vector machine
- data points
- probabilistic model
- information loss
- high quality
- similarity measure