Improved Winding Mechanical Fault Type Classification Methods Based on Polar Plots and Multiple Support Vector Machines.
Jiangnan LiuZhongyong ZhaoKai PangDong WangChao TangChenguo YaoPublished in: IEEE Access (2020)
Keyphrases
- pattern classification
- pattern recognition
- feature extraction
- support vector
- cross validation
- classification accuracy
- benchmark data sets
- feature vectors
- benchmark datasets
- machine learning
- classification method
- classification systems
- machine learning methods
- polar coordinates
- terms of classification accuracy
- feature ranking
- multiple types
- learning machines
- classification scheme
- generalization ability
- training set
- loss function
- support vector machine svm
- model selection
- image classification
- significant improvement