An Anonymization Method to Improve Data Utility for Classification.
Jianmin HanJuan YuJianfeng LuHao PengJiandang WuPublished in: CSS (2017)
Keyphrases
- synthetic data
- data sets
- training samples
- input data
- information loss
- pattern classification
- noisy data
- classification method
- test data
- classification algorithm
- missing data
- classification process
- support vector machine svm
- statistical methods
- support vector machine
- machine learning
- ordinal data
- input vectors
- cross validation
- extracted features
- bayesian methods
- feature selection
- data collection
- text classification
- classification accuracy
- pairwise
- data sources
- training data
- data structure
- statistical information
- preprocessing
- feature vectors
- prior knowledge
- missing values
- machine learning methods
- multi class
- decision rules
- original data
- active learning
- model selection
- data points
- classification trees
- data mining techniques
- clustering method