Modeling train timetables as images: A cost-sensitive deep learning framework for delay propagation pattern recognition.
Ping HuangZhongcan LiChao WenJavad LessanFrancesco CormanLiping FuPublished in: Expert Syst. Appl. (2021)
Keyphrases
- cost sensitive
- deep learning
- pattern recognition
- image classification
- multi class
- input image
- cost sensitive learning
- small number
- image features
- image processing
- machine learning
- neural network
- unsupervised learning
- naive bayes
- misclassification costs
- active learning
- object recognition
- test images
- data sets
- image segmentation
- class distribution