A deep transfer maximum classifier discrepancy method for rolling bearing fault diagnosis under few labeled data.
Zhenghong WuHongkai JiangTengfei LuKe ZhaoPublished in: Knowl. Based Syst. (2020)
Keyphrases
- fault diagnosis
- labeled data
- pairwise
- training data
- active learning
- semi supervised learning
- multi sensor information fusion
- supervised classifiers
- similarity measure
- supervised learning
- unsupervised learning
- bp neural network
- expert systems
- semi supervised
- monitoring and fault diagnosis
- supervised methods
- fault detection
- test data
- transfer learning
- data sets
- prior knowledge
- support vector machine
- training examples
- clustering method
- labeling process
- labeled instances
- artificial intelligence