Matches in SemOpenAlex for { <https://semopenalex.org/work/W2972193717> ?p ?o ?g. }
- W2972193717 abstract "Many learning tasks in the field of natural language processing including sequence tagging, sequence segmentation, and syntactic parsing have been successfully approached by means of structured prediction methods. An appealing property of the corresponding training algorithms is their ability to integrate the loss function of interest into the optimization process improving the final results according to the chosen measure of performance. Here, we focus on the task of constituency parsing and show how to optimize the model for the F1 -score in the max-margin framework of a structural support vector machine (SVM). For reasons of computational efficiency, it is a common approach to binarize the corresponding grammar before training. Unfortunately, this introduces a bias during the training procedure as the corresponding loss function is evaluated on the binary representation, while the resulting performance is measured on the original unbinarized trees. Here, we address this problem by extending the inference procedure presented by Bauer et al. Specifically, we propose an algorithmic modification that allows evaluating the loss on the unbinarized trees. The new approach properly models the loss function of interest resulting in better prediction accuracy and still benefits from the computational efficiency due to binarized representation. The presented idea can be easily transferred to other structured loss functions." @default.
- W2972193717 created "2019-09-12" @default.
- W2972193717 creator A5014060431 @default.
- W2972193717 creator A5034738514 @default.
- W2972193717 creator A5035494650 @default.
- W2972193717 creator A5072994165 @default.
- W2972193717 date "2019-01-01" @default.
- W2972193717 modified "2023-10-18" @default.
- W2972193717 title "Optimizing for Measure of Performance in Max-Margin Parsing" @default.
- W2972193717 cites W114565888 @default.
- W2972193717 cites W147273232 @default.
- W2972193717 cites W1506806321 @default.
- W2972193717 cites W1518470882 @default.
- W2972193717 cites W1551104980 @default.
- W2972193717 cites W1593793857 @default.
- W2972193717 cites W1632114991 @default.
- W2972193717 cites W1876515821 @default.
- W2972193717 cites W1970961429 @default.
- W2972193717 cites W1989926363 @default.
- W2972193717 cites W2008652694 @default.
- W2972193717 cites W2031248101 @default.
- W2972193717 cites W2069808690 @default.
- W2972193717 cites W2070771761 @default.
- W2972193717 cites W2092423930 @default.
- W2972193717 cites W2097606805 @default.
- W2972193717 cites W2097826433 @default.
- W2972193717 cites W2105636360 @default.
- W2972193717 cites W2105644991 @default.
- W2972193717 cites W2105842272 @default.
- W2972193717 cites W2111753845 @default.
- W2972193717 cites W2115364117 @default.
- W2972193717 cites W2116410915 @default.
- W2972193717 cites W2131962941 @default.
- W2972193717 cites W2142384583 @default.
- W2972193717 cites W2144414892 @default.
- W2972193717 cites W2182705849 @default.
- W2972193717 cites W220623884 @default.
- W2972193717 cites W2250993494 @default.
- W2972193717 cites W2343505200 @default.
- W2972193717 cites W2512425033 @default.
- W2972193717 cites W2564486991 @default.
- W2972193717 cites W2963084773 @default.
- W2972193717 cites W2963462075 @default.
- W2972193717 cites W2964262905 @default.
- W2972193717 doi "https://doi.org/10.1109/tnnls.2019.2934225" @default.
- W2972193717 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/31494564" @default.
- W2972193717 hasPublicationYear "2019" @default.
- W2972193717 type Work @default.
- W2972193717 sameAs 2972193717 @default.
- W2972193717 citedByCount "1" @default.
- W2972193717 countsByYear W29721937172019 @default.
- W2972193717 crossrefType "journal-article" @default.
- W2972193717 hasAuthorship W2972193717A5014060431 @default.
- W2972193717 hasAuthorship W2972193717A5034738514 @default.
- W2972193717 hasAuthorship W2972193717A5035494650 @default.
- W2972193717 hasAuthorship W2972193717A5072994165 @default.
- W2972193717 hasBestOaLocation W29721937172 @default.
- W2972193717 hasConcept C111472728 @default.
- W2972193717 hasConcept C111919701 @default.
- W2972193717 hasConcept C119857082 @default.
- W2972193717 hasConcept C12267149 @default.
- W2972193717 hasConcept C124101348 @default.
- W2972193717 hasConcept C138885662 @default.
- W2972193717 hasConcept C14036430 @default.
- W2972193717 hasConcept C153180895 @default.
- W2972193717 hasConcept C154945302 @default.
- W2972193717 hasConcept C162324750 @default.
- W2972193717 hasConcept C17744445 @default.
- W2972193717 hasConcept C186644900 @default.
- W2972193717 hasConcept C187736073 @default.
- W2972193717 hasConcept C189950617 @default.
- W2972193717 hasConcept C199539241 @default.
- W2972193717 hasConcept C204321447 @default.
- W2972193717 hasConcept C22367795 @default.
- W2972193717 hasConcept C2776214188 @default.
- W2972193717 hasConcept C2776359362 @default.
- W2972193717 hasConcept C2780009758 @default.
- W2972193717 hasConcept C2780451532 @default.
- W2972193717 hasConcept C41008148 @default.
- W2972193717 hasConcept C774472 @default.
- W2972193717 hasConcept C78458016 @default.
- W2972193717 hasConcept C86803240 @default.
- W2972193717 hasConcept C89600930 @default.
- W2972193717 hasConcept C94625758 @default.
- W2972193717 hasConcept C98045186 @default.
- W2972193717 hasConceptScore W2972193717C111472728 @default.
- W2972193717 hasConceptScore W2972193717C111919701 @default.
- W2972193717 hasConceptScore W2972193717C119857082 @default.
- W2972193717 hasConceptScore W2972193717C12267149 @default.
- W2972193717 hasConceptScore W2972193717C124101348 @default.
- W2972193717 hasConceptScore W2972193717C138885662 @default.
- W2972193717 hasConceptScore W2972193717C14036430 @default.
- W2972193717 hasConceptScore W2972193717C153180895 @default.
- W2972193717 hasConceptScore W2972193717C154945302 @default.
- W2972193717 hasConceptScore W2972193717C162324750 @default.
- W2972193717 hasConceptScore W2972193717C17744445 @default.
- W2972193717 hasConceptScore W2972193717C186644900 @default.
- W2972193717 hasConceptScore W2972193717C187736073 @default.
- W2972193717 hasConceptScore W2972193717C189950617 @default.
- W2972193717 hasConceptScore W2972193717C199539241 @default.