Matches in SemOpenAlex for { <https://semopenalex.org/work/W93205924> ?p ?o ?g. }
Showing items 1 to 89 of
89
with 100 items per page.
- W93205924 endingPage "809" @default.
- W93205924 startingPage "801" @default.
- W93205924 abstract "In Sequential Viterbi Models, such as HMMs, MEMMs, and Linear Chain CRFs, the type of patterns over output sequences that can be learned by the model depend directly on the model’s structure: any pattern that spans more output tags than are covered by the models’ order will be very difficult to learn. However, increasing a model’s order can lead to an increase in the number of model parameters, making the model more susceptible to sparse data problems. This paper shows how the notion of output transformation can be used to explore a variety of alternative model structures. Using output transformations, we can selectively increase the amount of contextual information available for some conditions, but not for others, thus allowing us to capture longer-distance consistencies while avoiding unnecessary increases to the model’s parameter space. The appropriate output transformation for a given task can be selected by applying a hill-climbing approach to heldout data. On the NP Chunking task, our hill-climbing system finds a model structure that outperformsbothfirst-orderand secondorder models with the same input feature set." @default.
- W93205924 created "2016-06-24" @default.
- W93205924 creator A5072285427 @default.
- W93205924 date "2007-06-01" @default.
- W93205924 modified "2023-09-26" @default.
- W93205924 title "Finding Good Sequential Model Structures using Output Transformations" @default.
- W93205924 cites W1564649749 @default.
- W93205924 cites W1623072288 @default.
- W93205924 cites W1667986067 @default.
- W93205924 cites W1766290689 @default.
- W93205924 cites W1934019294 @default.
- W93205924 cites W2008830554 @default.
- W93205924 cites W2125529971 @default.
- W93205924 cites W2156515921 @default.
- W93205924 cites W2162157640 @default.
- W93205924 cites W2162995740 @default.
- W93205924 hasPublicationYear "2007" @default.
- W93205924 type Work @default.
- W93205924 sameAs 93205924 @default.
- W93205924 citedByCount "1" @default.
- W93205924 crossrefType "proceedings-article" @default.
- W93205924 hasAuthorship W93205924A5072285427 @default.
- W93205924 hasConcept C104317684 @default.
- W93205924 hasConcept C11413529 @default.
- W93205924 hasConcept C119857082 @default.
- W93205924 hasConcept C138885662 @default.
- W93205924 hasConcept C152565575 @default.
- W93205924 hasConcept C154945302 @default.
- W93205924 hasConcept C162324750 @default.
- W93205924 hasConcept C177264268 @default.
- W93205924 hasConcept C185592680 @default.
- W93205924 hasConcept C187736073 @default.
- W93205924 hasConcept C199360897 @default.
- W93205924 hasConcept C203357204 @default.
- W93205924 hasConcept C204241405 @default.
- W93205924 hasConcept C2775953691 @default.
- W93205924 hasConcept C2776401178 @default.
- W93205924 hasConcept C2780451532 @default.
- W93205924 hasConcept C41008148 @default.
- W93205924 hasConcept C41895202 @default.
- W93205924 hasConcept C55493867 @default.
- W93205924 hasConceptScore W93205924C104317684 @default.
- W93205924 hasConceptScore W93205924C11413529 @default.
- W93205924 hasConceptScore W93205924C119857082 @default.
- W93205924 hasConceptScore W93205924C138885662 @default.
- W93205924 hasConceptScore W93205924C152565575 @default.
- W93205924 hasConceptScore W93205924C154945302 @default.
- W93205924 hasConceptScore W93205924C162324750 @default.
- W93205924 hasConceptScore W93205924C177264268 @default.
- W93205924 hasConceptScore W93205924C185592680 @default.
- W93205924 hasConceptScore W93205924C187736073 @default.
- W93205924 hasConceptScore W93205924C199360897 @default.
- W93205924 hasConceptScore W93205924C203357204 @default.
- W93205924 hasConceptScore W93205924C204241405 @default.
- W93205924 hasConceptScore W93205924C2775953691 @default.
- W93205924 hasConceptScore W93205924C2776401178 @default.
- W93205924 hasConceptScore W93205924C2780451532 @default.
- W93205924 hasConceptScore W93205924C41008148 @default.
- W93205924 hasConceptScore W93205924C41895202 @default.
- W93205924 hasConceptScore W93205924C55493867 @default.
- W93205924 hasLocation W932059241 @default.
- W93205924 hasOpenAccess W93205924 @default.
- W93205924 hasPrimaryLocation W932059241 @default.
- W93205924 hasRelatedWork W143558673 @default.
- W93205924 hasRelatedWork W1548974228 @default.
- W93205924 hasRelatedWork W2023340601 @default.
- W93205924 hasRelatedWork W2126763700 @default.
- W93205924 hasRelatedWork W2157140289 @default.
- W93205924 hasRelatedWork W2728766219 @default.
- W93205924 hasRelatedWork W2770331030 @default.
- W93205924 hasRelatedWork W2949923404 @default.
- W93205924 hasRelatedWork W2954973147 @default.
- W93205924 hasRelatedWork W2998192669 @default.
- W93205924 hasRelatedWork W3010729701 @default.
- W93205924 hasRelatedWork W3034396216 @default.
- W93205924 hasRelatedWork W3037853434 @default.
- W93205924 hasRelatedWork W3127413113 @default.
- W93205924 hasRelatedWork W3170718858 @default.
- W93205924 hasRelatedWork W63974246 @default.
- W93205924 hasRelatedWork W86969866 @default.
- W93205924 hasRelatedWork W986037092 @default.
- W93205924 hasRelatedWork W1639984653 @default.
- W93205924 hasRelatedWork W1837680610 @default.
- W93205924 isParatext "false" @default.
- W93205924 isRetracted "false" @default.
- W93205924 magId "93205924" @default.
- W93205924 workType "article" @default.