Matches in SemOpenAlex for { <https://semopenalex.org/work/W1979945719> ?p ?o ?g. }
Showing items 1 to 100 of
100
with 100 items per page.
- W1979945719 endingPage "194" @default.
- W1979945719 startingPage "159" @default.
- W1979945719 abstract "Data sparseness or overfitting is a serious problem in natural language processing employing machine learning methods. This is still true even for the maximum entropy (ME) method, whose flexible modeling capability has alleviated data sparseness more successfully than the other probabilistic models in many NLP tasks. Although we usually estimate the model so that it completely satisfies the equality constraints on feature expectations with the ME method, complete satisfaction leads to undesirable overfitting, especially for sparse features, since the constraints derived from a limited amount of training data are always uncertain. To control overfitting in ME estimation, we propose the use of box-type inequality constraints, where equality can be violated up to certain predefined levels that reflect this uncertainty. The derived models, inequality ME models, in effect have regularized estimation with L1 norm penalties of bounded parameters. Most importantly, this regularized estimation enables the model parameters to become sparse. This can be thought of as automatic feature selection, which is expected to improve generalization performance further. We evaluate the inequality ME models on text categorization datasets, and demonstrate their advantages over standard ME estimation, similarly motivated Gaussian MAP estimation of ME models, and support vector machines (SVMs), which are one of the state-of-the-art methods for text categorization." @default.
- W1979945719 created "2016-06-24" @default.
- W1979945719 creator A5040725471 @default.
- W1979945719 creator A5073653623 @default.
- W1979945719 date "2005-09-01" @default.
- W1979945719 modified "2023-10-16" @default.
- W1979945719 title "Maximum Entropy Models with Inequality Constraints: A Case Study on Text Categorization" @default.
- W1979945719 cites W1495058637 @default.
- W1979945719 cites W1563088657 @default.
- W1979945719 cites W1574862351 @default.
- W1979945719 cites W1592796124 @default.
- W1979945719 cites W1978394996 @default.
- W1979945719 cites W2001792610 @default.
- W1979945719 cites W2027875450 @default.
- W1979945719 cites W2039620585 @default.
- W1979945719 cites W2043772506 @default.
- W1979945719 cites W2075517337 @default.
- W1979945719 cites W2076467305 @default.
- W1979945719 cites W2098379588 @default.
- W1979945719 cites W2102667697 @default.
- W1979945719 cites W2135983473 @default.
- W1979945719 cites W2149684865 @default.
- W1979945719 cites W2156515921 @default.
- W1979945719 cites W2156909104 @default.
- W1979945719 cites W2160842254 @default.
- W1979945719 cites W2163306339 @default.
- W1979945719 cites W4245322178 @default.
- W1979945719 doi "https://doi.org/10.1007/s10994-005-0911-3" @default.
- W1979945719 hasPublicationYear "2005" @default.
- W1979945719 type Work @default.
- W1979945719 sameAs 1979945719 @default.
- W1979945719 citedByCount "48" @default.
- W1979945719 countsByYear W19799457192012 @default.
- W1979945719 countsByYear W19799457192013 @default.
- W1979945719 countsByYear W19799457192014 @default.
- W1979945719 countsByYear W19799457192015 @default.
- W1979945719 countsByYear W19799457192016 @default.
- W1979945719 countsByYear W19799457192017 @default.
- W1979945719 countsByYear W19799457192019 @default.
- W1979945719 countsByYear W19799457192020 @default.
- W1979945719 countsByYear W19799457192021 @default.
- W1979945719 countsByYear W19799457192022 @default.
- W1979945719 countsByYear W19799457192023 @default.
- W1979945719 crossrefType "journal-article" @default.
- W1979945719 hasAuthorship W1979945719A5040725471 @default.
- W1979945719 hasAuthorship W1979945719A5073653623 @default.
- W1979945719 hasBestOaLocation W19799457191 @default.
- W1979945719 hasConcept C106301342 @default.
- W1979945719 hasConcept C119857082 @default.
- W1979945719 hasConcept C121332964 @default.
- W1979945719 hasConcept C138885662 @default.
- W1979945719 hasConcept C148483581 @default.
- W1979945719 hasConcept C153180895 @default.
- W1979945719 hasConcept C154945302 @default.
- W1979945719 hasConcept C22019652 @default.
- W1979945719 hasConcept C2776401178 @default.
- W1979945719 hasConcept C33923547 @default.
- W1979945719 hasConcept C41008148 @default.
- W1979945719 hasConcept C41895202 @default.
- W1979945719 hasConcept C50644808 @default.
- W1979945719 hasConcept C62520636 @default.
- W1979945719 hasConcept C94124525 @default.
- W1979945719 hasConcept C9679016 @default.
- W1979945719 hasConceptScore W1979945719C106301342 @default.
- W1979945719 hasConceptScore W1979945719C119857082 @default.
- W1979945719 hasConceptScore W1979945719C121332964 @default.
- W1979945719 hasConceptScore W1979945719C138885662 @default.
- W1979945719 hasConceptScore W1979945719C148483581 @default.
- W1979945719 hasConceptScore W1979945719C153180895 @default.
- W1979945719 hasConceptScore W1979945719C154945302 @default.
- W1979945719 hasConceptScore W1979945719C22019652 @default.
- W1979945719 hasConceptScore W1979945719C2776401178 @default.
- W1979945719 hasConceptScore W1979945719C33923547 @default.
- W1979945719 hasConceptScore W1979945719C41008148 @default.
- W1979945719 hasConceptScore W1979945719C41895202 @default.
- W1979945719 hasConceptScore W1979945719C50644808 @default.
- W1979945719 hasConceptScore W1979945719C62520636 @default.
- W1979945719 hasConceptScore W1979945719C94124525 @default.
- W1979945719 hasConceptScore W1979945719C9679016 @default.
- W1979945719 hasIssue "1-3" @default.
- W1979945719 hasLocation W19799457191 @default.
- W1979945719 hasOpenAccess W1979945719 @default.
- W1979945719 hasPrimaryLocation W19799457191 @default.
- W1979945719 hasRelatedWork W2244814250 @default.
- W1979945719 hasRelatedWork W2363775966 @default.
- W1979945719 hasRelatedWork W2375828317 @default.
- W1979945719 hasRelatedWork W2409149208 @default.
- W1979945719 hasRelatedWork W2767651786 @default.
- W1979945719 hasRelatedWork W2989932438 @default.
- W1979945719 hasRelatedWork W3099765033 @default.
- W1979945719 hasRelatedWork W4210794429 @default.
- W1979945719 hasRelatedWork W4283732135 @default.
- W1979945719 hasRelatedWork W4385743487 @default.
- W1979945719 hasVolume "60" @default.
- W1979945719 isParatext "false" @default.
- W1979945719 isRetracted "false" @default.
- W1979945719 magId "1979945719" @default.
- W1979945719 workType "article" @default.