On one experimental optical network dataset, post-processing threshold adjustment improved failure-detection F1 by 15.3% over an unbalanced random forest baseline, more than any pre- or in-processing method tested.
Model and data-centric machine learn- ing algorithms to address data scarcity for failure iden- tification
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Pre-, In-, and Post-Processing Class Imbalance Mitigation Techniques for Failure Detection in Optical Networks
On one experimental optical network dataset, post-processing threshold adjustment improved failure-detection F1 by 15.3% over an unbalanced random forest baseline, more than any pre- or in-processing method tested.