Matches in SemOpenAlex for { <https://semopenalex.org/work/W2018077356> ?p ?o ?g. }
Showing items 1 to 89 of
89
with 100 items per page.
- W2018077356 abstract "Well-calibrated probabilities are necessary in many applications like probabilistic frameworks or cost-sensitive tasks. Based on previous success of asymmetric Laplace method in calibrating text classifiers' scores, we propose to use piecewise logistic regression, which is a simple extension of standard logistic regression, as an alternative method in the discriminative family. We show that both methods have the flexibility to be piecewise linear functions in log-odds, but they are based on quite different assumptions. We evaluated asymmetric Laplace method, piecewise logistic regression and standard logistic regression over standard text categorization collections (Reuters-21578 and TRECAP) with three classifiers (SVM, Naive Bayes and Logistic Regression Classifier), and observed that piecewise logistic regression performs significantly better than the other two methods in the log-loss metric." @default.
- W2018077356 created "2016-06-24" @default.
- W2018077356 creator A5041434096 @default.
- W2018077356 creator A5057209439 @default.
- W2018077356 date "2004-01-01" @default.
- W2018077356 modified "2023-10-16" @default.
- W2018077356 title "Probabilistic score estimation with piecewise logistic regression" @default.
- W2018077356 cites W109189503 @default.
- W2018077356 cites W1507029541 @default.
- W2018077356 cites W1524688041 @default.
- W2018077356 cites W1550206324 @default.
- W2018077356 cites W1598033630 @default.
- W2018077356 cites W1602344127 @default.
- W2018077356 cites W1618905105 @default.
- W2018077356 cites W2012942264 @default.
- W2018077356 cites W2060216474 @default.
- W2018077356 cites W2085989833 @default.
- W2018077356 cites W2089086886 @default.
- W2018077356 cites W2149684865 @default.
- W2018077356 cites W2156909104 @default.
- W2018077356 cites W2163614729 @default.
- W2018077356 cites W2170768058 @default.
- W2018077356 cites W2435251607 @default.
- W2018077356 cites W2799061466 @default.
- W2018077356 cites W568726942 @default.
- W2018077356 doi "https://doi.org/10.1145/1015330.1015335" @default.
- W2018077356 hasPublicationYear "2004" @default.
- W2018077356 type Work @default.
- W2018077356 sameAs 2018077356 @default.
- W2018077356 citedByCount "20" @default.
- W2018077356 countsByYear W20180773562012 @default.
- W2018077356 countsByYear W20180773562020 @default.
- W2018077356 countsByYear W20180773562021 @default.
- W2018077356 countsByYear W20180773562022 @default.
- W2018077356 countsByYear W20180773562023 @default.
- W2018077356 crossrefType "proceedings-article" @default.
- W2018077356 hasAuthorship W2018077356A5041434096 @default.
- W2018077356 hasAuthorship W2018077356A5057209439 @default.
- W2018077356 hasConcept C105795698 @default.
- W2018077356 hasConcept C119857082 @default.
- W2018077356 hasConcept C120068334 @default.
- W2018077356 hasConcept C127413603 @default.
- W2018077356 hasConcept C134306372 @default.
- W2018077356 hasConcept C151956035 @default.
- W2018077356 hasConcept C152877465 @default.
- W2018077356 hasConcept C154945302 @default.
- W2018077356 hasConcept C164660894 @default.
- W2018077356 hasConcept C201995342 @default.
- W2018077356 hasConcept C33923547 @default.
- W2018077356 hasConcept C35519122 @default.
- W2018077356 hasConcept C41008148 @default.
- W2018077356 hasConcept C49937458 @default.
- W2018077356 hasConcept C61722155 @default.
- W2018077356 hasConcept C83546350 @default.
- W2018077356 hasConcept C96250715 @default.
- W2018077356 hasConceptScore W2018077356C105795698 @default.
- W2018077356 hasConceptScore W2018077356C119857082 @default.
- W2018077356 hasConceptScore W2018077356C120068334 @default.
- W2018077356 hasConceptScore W2018077356C127413603 @default.
- W2018077356 hasConceptScore W2018077356C134306372 @default.
- W2018077356 hasConceptScore W2018077356C151956035 @default.
- W2018077356 hasConceptScore W2018077356C152877465 @default.
- W2018077356 hasConceptScore W2018077356C154945302 @default.
- W2018077356 hasConceptScore W2018077356C164660894 @default.
- W2018077356 hasConceptScore W2018077356C201995342 @default.
- W2018077356 hasConceptScore W2018077356C33923547 @default.
- W2018077356 hasConceptScore W2018077356C35519122 @default.
- W2018077356 hasConceptScore W2018077356C41008148 @default.
- W2018077356 hasConceptScore W2018077356C49937458 @default.
- W2018077356 hasConceptScore W2018077356C61722155 @default.
- W2018077356 hasConceptScore W2018077356C83546350 @default.
- W2018077356 hasConceptScore W2018077356C96250715 @default.
- W2018077356 hasLocation W20180773561 @default.
- W2018077356 hasOpenAccess W2018077356 @default.
- W2018077356 hasPrimaryLocation W20180773561 @default.
- W2018077356 hasRelatedWork W1621861229 @default.
- W2018077356 hasRelatedWork W2016216486 @default.
- W2018077356 hasRelatedWork W2086639915 @default.
- W2018077356 hasRelatedWork W2127072394 @default.
- W2018077356 hasRelatedWork W2171183097 @default.
- W2018077356 hasRelatedWork W2362208065 @default.
- W2018077356 hasRelatedWork W2375721435 @default.
- W2018077356 hasRelatedWork W2750047195 @default.
- W2018077356 hasRelatedWork W39561353 @default.
- W2018077356 hasRelatedWork W2184922845 @default.
- W2018077356 isParatext "false" @default.
- W2018077356 isRetracted "false" @default.
- W2018077356 magId "2018077356" @default.
- W2018077356 workType "article" @default.