Semi-supervised deep learning for recognizing construction activity types from vibration monitoring data

Keyphrases

• 1-dimensional Convolutional Neural Network (1D-CNN) 75%
• Accelerometer Data 25%
• Activity Type 100%
• Adverse Impact 25%
• Automatic Recognition 25%
• Construction Activities 100%
• Construction Site 25%
• Construction Vibration 50%
• Convolutional Neural Network 75%
• Convolutional Neural Network Model 25%
• Deep Learning Algorithm 25%
• Labeled Data 75%
• Ladder Network 25%
• Manual Identification 25%
• Monitoring Data 100%
• Noise Level 25%
• Pseudo Label 50%
• Raw Acceleration 25%
• Semi-supervised Algorithm 25%
• Semi-supervised Deep Learning 100%
• Supervised Deep Learning 25%
• Unlabeled Data 25%
• Vibration Data 50%
• Vibration Database 25%
• Vibration Monitoring 100%
• Vibration Prediction 25%

Engineering

• Construction Activity 100%
• Construction Monitoring 16%
• Construction Site 16%
• Convolutional Neural Network 100%
• Data Collection 16%
• Deep Learning Method 100%
• Empirical Model 16%
• Induced Vibration 33%
• Ladder Networks 16%
• Measured Data 16%
• Monitoring Data 100%
• Network Model 16%
• Noise Level 16%
• One Dimensional 50%
• Vibration Condition Monitoring 100%

Computer Science

• Adverse Impact 16%
• Construction Site 16%
• Convolutional Neural Network 100%
• Data Collection 16%
• Deep Learning Method 100%
• Induced Vibration 33%
• Monitoring Data 100%
• Neural Network Model 16%
• Unlabeled Data 16%