Improving accuracy of classification models induced from anonymized datasets

Business & Economics

• Anonymization 100%
• Classifier 74%
• Repository 46%
• Data Mining 30%
• Privacy Preserving 29%
• Miners 26%
• Anonymity 24%
• Public Organizations 21%
• Public-private 20%
• Experiment 12%
• Performance 8%

Mathematics

• Classifier 52%
• Repository 44%
• Privacy Preserving 24%
• Information Loss 22%
• Classification Algorithm 22%
• Anonymity 22%
• Model 20%
• Value Distribution 19%
• Data Mining 18%
• Attribute 15%
• Experiment 11%
• Tend 11%
• Performance 9%
• Generalization 7%
• Form 6%

Engineering & Materials Science

• Classifiers 44%
• Miners 24%
• Data privacy 21%
• Data mining 17%
• Experiments 7%