Add noise to remove noise: Local differential privacy for feature selection

Keyphrases

• Feature Selection 100%
• Remove Noise 100%
• Local Differential Privacy 100%
• User Device 20%
• Prediction Accuracy 20%
• World Wide Web 20%
• Data Provider 20%
• Confidentiality 20%
• Overfitting 20%
• Modernity 20%
• Informative Features 20%
• Discriminating Features 20%
• Dataset Properties 20%
• Protected Data 20%
• Feature Selection Methods 20%
• Privacy Sensitive 20%
• Selected Subset 20%
• Selection Framework 20%
• Level Selection 20%
• Privacy Mechanisms 20%
• Trusted Party 20%
• Privacy Features 20%
• Privacy Level 20%

Computer Science

• Feature Selection 100%
• Differential Privacy 100%
• Feature Extraction 100%
• Experimental Result 16%
• Practical Solution 16%
• Provider Data 16%
• Underlying Data 16%
• Privacy Property 16%