In multi-label rule learning, the best consistency-coverage tradeoff of the scoring heuristic depends on the target measure, and locally optimal rules do not always yield globally optimal models.
Gibaja, and Sebasti´ an Ventura
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
On the Trade-off Between Consistency and Coverage in Multi-label Rule Learning Heuristics
In multi-label rule learning, the best consistency-coverage tradeoff of the scoring heuristic depends on the target measure, and locally optimal rules do not always yield globally optimal models.