A Cost-Sensitive Shared Hidden Layer Autoencoder for Cross-Project Defect Prediction.
Juanjuan LiXiao-Yuan JingFei WuYing SunYongguang YangPublished in: PRCV (3) (2019)
Keyphrases
- cost sensitive
- hidden layer
- defect prediction
- output layer
- feedforward neural networks
- back propagation
- feed forward
- software projects
- learning rate
- multi class
- misclassification costs
- multilayer perceptron
- neural network
- artificial neural networks
- activation function
- recurrent neural networks
- radial basis function
- neural nets
- rbf neural network
- active learning
- project management
- latent variables
- training error
- naive bayes
- support vector machine
- boosting algorithms
- software repositories
- source code
- learning algorithm