HIGHLIGHTS
SUMMARY
Classification-based with training on one or more classes (e_g, COMPOSE ); Clustering-based with training on one or more classes (e_g, OLINDDA, MINAS ); Clustering-based that can be applied from scratch (e_g, ADP ). The main paper innovation consists of the introduction of a supervised neural_network that analyzes the incoming time series, classifying the type of fault in a supervised manner. The authors provide two implementations for the Classifier, based on a supervised and an neural_network that the authors train in the implementation on the series generated by the encoder unsupervised approach, respectively. The unsupervised . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.