A methodology is proposed to detect class-based concept drift by self-evaluating predictive model degradation, with experiments on synthetic and real-world datasets showing effectiveness.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Automating concept-drift detection by self-evaluating predictive model degradation
A methodology is proposed to detect class-based concept drift by self-evaluating predictive model degradation, with experiments on synthetic and real-world datasets showing effectiveness.