Deep learning approach for defective spot welds classification using small and class-imbalanced datasets.
Wei DaiDayong LiDing TangHuamiao WangYinghong PengPublished in: Neurocomputing (2022)
Keyphrases
- imbalanced datasets
- deep learning
- cost sensitive learning
- class imbalance
- class distribution
- imbalanced data
- decision trees
- minority class
- class labels
- ensemble methods
- sampling methods
- training dataset
- machine learning
- unsupervised learning
- classification algorithm
- pattern recognition
- multi class classification
- high dimensionality
- machine learning methods
- benchmark datasets
- support vector machine
- classification accuracy
- feature vectors
- feature extraction
- cost sensitive
- feature selection algorithms
- image classification
- probability estimation
- training set
- learning algorithm