A Novel Multiview Sampling-based Meta Self-Paced Learning Approach for Class-imbalanced Intelligent Fault Diagnosis

Keyphrases

• Sampling Methods 100%
• Learning Approaches 100%
• Class Imbalance 100%
• Intelligent Fault Diagnosis 100%
• Self-paced Learning 100%
• Fault Diagnosis Model 66%
• Balanced Dataset 66%
• Spaced Learning 66%
• Signal Processing Techniques 33%
• Statistical Features 33%
• Machine Learning Models 33%
• Time-frequency Domain 33%
• Service Applications 33%
• Value Loss 33%
• Feature Data 33%
• Noise Sample 33%
• Fault Condition 33%
• One-based 33%
• Diagnostic Accuracy 33%
• Learning Technologies 33%
• Sampling Process 33%
• Convolutional Neural Network Model 33%
• Fault Diagnosis 33%
• Sample Quality 33%
• Public Dataset 33%
• Motor Tests 33%
• Overgeneralization 33%
• Multi-view Features 33%
• Class-imbalanced Datasets 33%
• Unbalanced Faults 33%
• High-quality Sample 33%
• Setting Parameters 33%
• Machine Fault Diagnosis 33%
• Industrial Motor 33%
• Large Loss 33%
• Practical Machine 33%
• Traditional Machine Learning 33%
• Modular Services 33%

Computer Science

• Learning Approach 100%
• Fault Diagnosis 100%
• Intelligent Fault Diagnosis 100%
• Frequency Domain 50%
• Neural Network Model 25%
• Machine Learning 25%
• Learning System 25%
• Experimental Result 25%
• Service Application 25%
• Statistical Feature 25%
• Baseline Method 25%
• Learning Technology 25%
• Processing Method 25%
• Class Imbalance 25%
• Sampling Process 25%
• Convolutional Neural Network 25%

Engineering

• Learning Approach 100%
• Fault Diagnosis 100%
• Frequency Domain 40%
• Processing Method 20%
• Experimental Result 20%
• Network Model 20%
• Time Domain 20%
• Level Model 20%
• Learning System 20%
• Statistical Feature 20%
• Sampling Process 20%
• Convolutional Neural Network 20%
• Data Sample 20%