A similarity-and-loss threshold rule decides when to update prediction models on industrial data streams, reported to cut RMSE by about 64% and raise AUC from 0.68 to 0.91 on two power-plant datasets.
Prognostic modelling options for remaining useful life estimation by industry,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2019 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
A General Data Renewal Model for Prediction Algorithms in Industrial Data Analytics
A similarity-and-loss threshold rule decides when to update prediction models on industrial data streams, reported to cut RMSE by about 64% and raise AUC from 0.68 to 0.91 on two power-plant datasets.