FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W28749054> ?p ?o ?g. }

• W28749054 endingPage "1945" @default.
• W28749054 startingPage "1939" @default.
• W28749054 abstract "In classification problems, isotonic regression has been commonly used to map the prediction scores to posterior class probabilities. However, isotonic regression may suffer from overfitting, and the learned mapping is often discontinuous. Besides, current efforts mainly focus on the calibration of a single classifier. As different classifiers have different strengths, a combination of them can lead to better performance. In this paper, we propose a novel probability calibration approach for such an ensemble of classifiers. We first construct isotonic constraints on the desired probabilities based on soft voting of the classifiers. Manifold information is also incorporated to combat overfitting and ensure function smoothness. Computationally, the extended isotonic regression model can be learned efficiently by a novel optimization algorithm based on the alternating direction method of multipliers (ADMM). Experiments on a number of real-world data sets demonstrate that the proposed approach consistently outperforms independent classifiers and other combinations of the classifiers' probabilities in terms of the Brier score and AUC." @default.
• W28749054 created "2016-06-24" @default.
• W28749054 creator A5070273088 @default.
• W28749054 creator A5080906311 @default.
• W28749054 date "2013-08-03" @default.
• W28749054 modified "2023-09-26" @default.
• W28749054 title "Accurate probability calibration for multiple classifiers" @default.
• W28749054 cites W1489122873 @default.
• W28749054 cites W1507039213 @default.
• W28749054 cites W1536484404 @default.
• W28749054 cites W1549918636 @default.
• W28749054 cites W1618905105 @default.
• W28749054 cites W1977245551 @default.
• W28749054 cites W1979713464 @default.
• W28749054 cites W1998128748 @default.
• W28749054 cites W2006560229 @default.
• W28749054 cites W2012942264 @default.
• W28749054 cites W2029896651 @default.
• W28749054 cites W2073241381 @default.
• W28749054 cites W2090883204 @default.
• W28749054 cites W2098824882 @default.
• W28749054 cites W2104290444 @default.
• W28749054 cites W2104452301 @default.
• W28749054 cites W2107103101 @default.
• W28749054 cites W2118585731 @default.
• W28749054 cites W2123157758 @default.
• W28749054 cites W2128302979 @default.
• W28749054 cites W2129018774 @default.
• W28749054 cites W2142082830 @default.
• W28749054 cites W2158698691 @default.
• W28749054 cites W2164278908 @default.
• W28749054 cites W2165664908 @default.
• W28749054 cites W2166290812 @default.
• W28749054 cites W2244987555 @default.
• W28749054 cites W3145128584 @default.
• W28749054 cites W605727707 @default.
• W28749054 hasPublicationYear "2013" @default.
• W28749054 type Work @default.
• W28749054 sameAs 28749054 @default.
• W28749054 citedByCount "17" @default.
• W28749054 countsByYear W287490542015 @default.
• W28749054 countsByYear W287490542016 @default.
• W28749054 countsByYear W287490542017 @default.
• W28749054 countsByYear W287490542018 @default.
• W28749054 countsByYear W287490542020 @default.
• W28749054 countsByYear W287490542021 @default.
• W28749054 crossrefType "proceedings-article" @default.
• W28749054 hasAuthorship W28749054A5070273088 @default.
• W28749054 hasAuthorship W28749054A5080906311 @default.
• W28749054 hasConcept C105795698 @default.
• W28749054 hasConcept C106135958 @default.
• W28749054 hasConcept C119857082 @default.
• W28749054 hasConcept C153180895 @default.
• W28749054 hasConcept C154945302 @default.
• W28749054 hasConcept C165838908 @default.
• W28749054 hasConcept C17418463 @default.
• W28749054 hasConcept C185429906 @default.
• W28749054 hasConcept C22019652 @default.
• W28749054 hasConcept C33923547 @default.
• W28749054 hasConcept C35405484 @default.
• W28749054 hasConcept C41008148 @default.
• W28749054 hasConcept C50644808 @default.
• W28749054 hasConcept C81388566 @default.
• W28749054 hasConcept C83546350 @default.
• W28749054 hasConcept C95623464 @default.
• W28749054 hasConceptScore W28749054C105795698 @default.
• W28749054 hasConceptScore W28749054C106135958 @default.
• W28749054 hasConceptScore W28749054C119857082 @default.
• W28749054 hasConceptScore W28749054C153180895 @default.
• W28749054 hasConceptScore W28749054C154945302 @default.
• W28749054 hasConceptScore W28749054C165838908 @default.
• W28749054 hasConceptScore W28749054C17418463 @default.
• W28749054 hasConceptScore W28749054C185429906 @default.
• W28749054 hasConceptScore W28749054C22019652 @default.
• W28749054 hasConceptScore W28749054C33923547 @default.
• W28749054 hasConceptScore W28749054C35405484 @default.
• W28749054 hasConceptScore W28749054C41008148 @default.
• W28749054 hasConceptScore W28749054C50644808 @default.
• W28749054 hasConceptScore W28749054C81388566 @default.
• W28749054 hasConceptScore W28749054C83546350 @default.
• W28749054 hasConceptScore W28749054C95623464 @default.
• W28749054 hasLocation W287490541 @default.
• W28749054 hasOpenAccess W28749054 @default.
• W28749054 hasPrimaryLocation W287490541 @default.
• W28749054 hasRelatedWork W1542959211 @default.
• W28749054 hasRelatedWork W1565746575 @default.
• W28749054 hasRelatedWork W1598033630 @default.
• W28749054 hasRelatedWork W1599263113 @default.
• W28749054 hasRelatedWork W1602946365 @default.
• W28749054 hasRelatedWork W1618905105 @default.
• W28749054 hasRelatedWork W1965755064 @default.
• W28749054 hasRelatedWork W2012942264 @default.
• W28749054 hasRelatedWork W2068618911 @default.
• W28749054 hasRelatedWork W2098824882 @default.
• W28749054 hasRelatedWork W2107103101 @default.
• W28749054 hasRelatedWork W2137556846 @default.
• W28749054 hasRelatedWork W2153635508 @default.
• W28749054 hasRelatedWork W2170112109 @default.

• next