A data anonymous method based on overlapping slicing.
Jing YangZiyun LiuYue YangJianpei ZhangPublished in: CSCWD (2014)
Keyphrases
- synthetic data
- data sets
- raw data
- test data
- detection method
- input data
- prior knowledge
- dynamic programming
- high quality
- data processing
- data sources
- database
- statistical methods
- high accuracy
- knowledge discovery
- data analysis
- missing data
- training data
- similarity measure
- noisy data
- cost function
- correlation analysis
- missing values
- statistical analysis
- statistical significance
- data structure
- support vector machine
- image data
- significant improvement
- training samples
- em algorithm
- segmentation method
- data mining techniques
- high precision
- data points
- original data
- prior information
- spectral clustering
- computational cost
- information loss
- high dimensional