FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2945100890> ?p ?o ?g. }

• W2945100890 abstract "The uncertainty measurement of classifiers' predictions is especially important in applications such as medical diagnoses that need to ensure limited human resources can focus on the most uncertain predictions returned by machine learning models. However, few existing uncertainty models attempt to improve overall prediction accuracy where human resources are involved in the text classification task. In this paper, we propose a novel neural-network-based model that applies a new dropout-entropy method for uncertainty measurement. We also design a metric learning method on feature representations, which can boost the performance of dropout-based uncertainty methods with smaller prediction variance in accurate prediction trials. Extensive experiments on real-world data sets demonstrate that our method can achieve a considerable improvement in overall prediction accuracy compared to existing approaches. In particular, our model improved the accuracy from 0.78 to 0.92 when 30% of the most uncertain predictions were handed over to human experts in 20NewsGroup data." @default.
• W2945100890 created "2019-05-29" @default.
• W2945100890 creator A5035052603 @default.
• W2945100890 creator A5038002204 @default.
• W2945100890 creator A5052809293 @default.
• W2945100890 creator A5074151953 @default.
• W2945100890 date "2019-07-17" @default.
• W2945100890 modified "2023-09-25" @default.
• W2945100890 title "Mitigating Uncertainty in Document Classification" @default.
• W2945100890 cites W132090532 @default.
• W2945100890 cites W1488642921 @default.
• W2945100890 cites W1493526108 @default.
• W2945100890 cites W1502922572 @default.
• W2945100890 cites W1522301498 @default.
• W2945100890 cites W1719489212 @default.
• W2945100890 cites W2059993991 @default.
• W2945100890 cites W2061873838 @default.
• W2945100890 cites W2070813603 @default.
• W2945100890 cites W2095705004 @default.
• W2945100890 cites W2106053110 @default.
• W2945100890 cites W2109824782 @default.
• W2945100890 cites W2113459411 @default.
• W2945100890 cites W2117154949 @default.
• W2945100890 cites W2127538960 @default.
• W2945100890 cites W2130237711 @default.
• W2945100890 cites W2130556178 @default.
• W2945100890 cites W2171585602 @default.
• W2945100890 cites W2187089797 @default.
• W2945100890 cites W2250539671 @default.
• W2945100890 cites W2325939864 @default.
• W2945100890 cites W2338752163 @default.
• W2945100890 cites W2470673105 @default.
• W2945100890 cites W2604433096 @default.
• W2945100890 cites W2622942114 @default.
• W2945100890 cites W2740379828 @default.
• W2945100890 cites W2758425594 @default.
• W2945100890 cites W2766678531 @default.
• W2945100890 cites W2950517871 @default.
• W2945100890 cites W2963207607 @default.
• W2945100890 cites W2963238274 @default.
• W2945100890 cites W2964059111 @default.
• W2945100890 cites W2964308564 @default.
• W2945100890 cites W3040426824 @default.
• W2945100890 doi "https://doi.org/10.48550/arxiv.1907.07590" @default.
• W2945100890 hasPublicationYear "2019" @default.
• W2945100890 type Work @default.
• W2945100890 sameAs 2945100890 @default.
• W2945100890 citedByCount "4" @default.
• W2945100890 countsByYear W29451008902019 @default.
• W2945100890 countsByYear W29451008902020 @default.
• W2945100890 countsByYear W29451008902021 @default.
• W2945100890 crossrefType "posted-content" @default.
• W2945100890 hasAuthorship W2945100890A5035052603 @default.
• W2945100890 hasAuthorship W2945100890A5038002204 @default.
• W2945100890 hasAuthorship W2945100890A5052809293 @default.
• W2945100890 hasAuthorship W2945100890A5074151953 @default.
• W2945100890 hasBestOaLocation W29451008901 @default.
• W2945100890 hasConcept C106301342 @default.
• W2945100890 hasConcept C119857082 @default.
• W2945100890 hasConcept C121332964 @default.
• W2945100890 hasConcept C121955636 @default.
• W2945100890 hasConcept C124101348 @default.
• W2945100890 hasConcept C127413603 @default.
• W2945100890 hasConcept C138885662 @default.
• W2945100890 hasConcept C142724271 @default.
• W2945100890 hasConcept C144133560 @default.
• W2945100890 hasConcept C154945302 @default.
• W2945100890 hasConcept C176217482 @default.
• W2945100890 hasConcept C196083921 @default.
• W2945100890 hasConcept C201995342 @default.
• W2945100890 hasConcept C21547014 @default.
• W2945100890 hasConcept C2776145597 @default.
• W2945100890 hasConcept C2776401178 @default.
• W2945100890 hasConcept C2780451532 @default.
• W2945100890 hasConcept C41008148 @default.
• W2945100890 hasConcept C41895202 @default.
• W2945100890 hasConcept C50644808 @default.
• W2945100890 hasConcept C534262118 @default.
• W2945100890 hasConcept C62520636 @default.
• W2945100890 hasConcept C71924100 @default.
• W2945100890 hasConceptScore W2945100890C106301342 @default.
• W2945100890 hasConceptScore W2945100890C119857082 @default.
• W2945100890 hasConceptScore W2945100890C121332964 @default.
• W2945100890 hasConceptScore W2945100890C121955636 @default.
• W2945100890 hasConceptScore W2945100890C124101348 @default.
• W2945100890 hasConceptScore W2945100890C127413603 @default.
• W2945100890 hasConceptScore W2945100890C138885662 @default.
• W2945100890 hasConceptScore W2945100890C142724271 @default.
• W2945100890 hasConceptScore W2945100890C144133560 @default.
• W2945100890 hasConceptScore W2945100890C154945302 @default.
• W2945100890 hasConceptScore W2945100890C176217482 @default.
• W2945100890 hasConceptScore W2945100890C196083921 @default.
• W2945100890 hasConceptScore W2945100890C201995342 @default.
• W2945100890 hasConceptScore W2945100890C21547014 @default.
• W2945100890 hasConceptScore W2945100890C2776145597 @default.
• W2945100890 hasConceptScore W2945100890C2776401178 @default.
• W2945100890 hasConceptScore W2945100890C2780451532 @default.
• W2945100890 hasConceptScore W2945100890C41008148 @default.
• W2945100890 hasConceptScore W2945100890C41895202 @default.
• W2945100890 hasConceptScore W2945100890C50644808 @default.

• next