Matches in SemOpenAlex for { <https://semopenalex.org/work/W2158717066> ?p ?o ?g. }
- W2158717066 endingPage "2939" @default.
- W2158717066 startingPage "2931" @default.
- W2158717066 abstract "This paper addresses the problem of unsupervised feature learning for text data. Our method is grounded in the principle of minimum description length and uses a dictionary-based compression scheme to extract a succinct feature set. Specifically, our method finds a set of word k-grams that minimizes the cost of reconstructing the text losslessly. We formulate document compression as a binary optimization task and show how to solve it approximately via a sequence of reweighted linear programs that are efficient to solve and parallelizable. As our method is unsupervised, features may be extracted once and subsequently used in a variety of tasks. We demonstrate the performance of these features over a range of scenarios including unsupervised exploratory analysis and supervised text categorization. Our compressed feature space is two orders of magnitude smaller than the full k-gram space and matches the text categorization accuracy achieved in the full feature space. This dimensionality reduction not only results in faster training times, but it can also help elucidate structure in unsupervised learning tasks and reduce the amount of training data necessary for supervised learning." @default.
- W2158717066 created "2016-06-24" @default.
- W2158717066 creator A5025508471 @default.
- W2158717066 creator A5044981994 @default.
- W2158717066 creator A5045744709 @default.
- W2158717066 creator A5060775456 @default.
- W2158717066 date "2013-12-05" @default.
- W2158717066 modified "2023-09-24" @default.
- W2158717066 title "Compressive Feature Learning" @default.
- W2158717066 cites W1487016832 @default.
- W2158717066 cites W1493526108 @default.
- W2158717066 cites W1748436461 @default.
- W2158717066 cites W1964155876 @default.
- W2158717066 cites W1975900269 @default.
- W2158717066 cites W1978119584 @default.
- W2158717066 cites W2028912194 @default.
- W2158717066 cites W2050559027 @default.
- W2158717066 cites W2054658115 @default.
- W2158717066 cites W2072994259 @default.
- W2158717066 cites W2097360283 @default.
- W2158717066 cites W2102831150 @default.
- W2158717066 cites W2104290684 @default.
- W2158717066 cites W2107745473 @default.
- W2158717066 cites W2107861471 @default.
- W2158717066 cites W2113459411 @default.
- W2158717066 cites W2116237179 @default.
- W2158717066 cites W2119479037 @default.
- W2158717066 cites W2124858543 @default.
- W2158717066 cites W2128859735 @default.
- W2158717066 cites W2130698119 @default.
- W2158717066 cites W2135046866 @default.
- W2158717066 cites W2164278908 @default.
- W2158717066 cites W2165340347 @default.
- W2158717066 cites W2166064672 @default.
- W2158717066 cites W2171886309 @default.
- W2158717066 cites W2435251607 @default.
- W2158717066 hasPublicationYear "2013" @default.
- W2158717066 type Work @default.
- W2158717066 sameAs 2158717066 @default.
- W2158717066 citedByCount "11" @default.
- W2158717066 countsByYear W21587170662014 @default.
- W2158717066 countsByYear W21587170662015 @default.
- W2158717066 countsByYear W21587170662016 @default.
- W2158717066 countsByYear W21587170662018 @default.
- W2158717066 countsByYear W21587170662019 @default.
- W2158717066 countsByYear W21587170662020 @default.
- W2158717066 crossrefType "proceedings-article" @default.
- W2158717066 hasAuthorship W2158717066A5025508471 @default.
- W2158717066 hasAuthorship W2158717066A5044981994 @default.
- W2158717066 hasAuthorship W2158717066A5045744709 @default.
- W2158717066 hasAuthorship W2158717066A5060775456 @default.
- W2158717066 hasConcept C111030470 @default.
- W2158717066 hasConcept C119857082 @default.
- W2158717066 hasConcept C138885662 @default.
- W2158717066 hasConcept C153180895 @default.
- W2158717066 hasConcept C154945302 @default.
- W2158717066 hasConcept C177264268 @default.
- W2158717066 hasConcept C199360897 @default.
- W2158717066 hasConcept C2524010 @default.
- W2158717066 hasConcept C2776401178 @default.
- W2158717066 hasConcept C33923547 @default.
- W2158717066 hasConcept C41008148 @default.
- W2158717066 hasConcept C41895202 @default.
- W2158717066 hasConcept C52622490 @default.
- W2158717066 hasConcept C70518039 @default.
- W2158717066 hasConcept C78548338 @default.
- W2158717066 hasConcept C8038995 @default.
- W2158717066 hasConcept C83665646 @default.
- W2158717066 hasConcept C90805587 @default.
- W2158717066 hasConceptScore W2158717066C111030470 @default.
- W2158717066 hasConceptScore W2158717066C119857082 @default.
- W2158717066 hasConceptScore W2158717066C138885662 @default.
- W2158717066 hasConceptScore W2158717066C153180895 @default.
- W2158717066 hasConceptScore W2158717066C154945302 @default.
- W2158717066 hasConceptScore W2158717066C177264268 @default.
- W2158717066 hasConceptScore W2158717066C199360897 @default.
- W2158717066 hasConceptScore W2158717066C2524010 @default.
- W2158717066 hasConceptScore W2158717066C2776401178 @default.
- W2158717066 hasConceptScore W2158717066C33923547 @default.
- W2158717066 hasConceptScore W2158717066C41008148 @default.
- W2158717066 hasConceptScore W2158717066C41895202 @default.
- W2158717066 hasConceptScore W2158717066C52622490 @default.
- W2158717066 hasConceptScore W2158717066C70518039 @default.
- W2158717066 hasConceptScore W2158717066C78548338 @default.
- W2158717066 hasConceptScore W2158717066C8038995 @default.
- W2158717066 hasConceptScore W2158717066C83665646 @default.
- W2158717066 hasConceptScore W2158717066C90805587 @default.
- W2158717066 hasLocation W21587170661 @default.
- W2158717066 hasOpenAccess W2158717066 @default.
- W2158717066 hasPrimaryLocation W21587170661 @default.
- W2158717066 hasRelatedWork W2113459411 @default.
- W2158717066 hasRelatedWork W2135046866 @default.
- W2158717066 hasRelatedWork W2283332108 @default.
- W2158717066 hasRelatedWork W2394168373 @default.
- W2158717066 hasRelatedWork W2407332982 @default.
- W2158717066 hasRelatedWork W2409455427 @default.
- W2158717066 hasRelatedWork W2422268042 @default.
- W2158717066 hasRelatedWork W2527508530 @default.