Matches in SemOpenAlex for { <https://semopenalex.org/work/W234712689> ?p ?o ?g. }
Showing items 1 to 80 of
80
with 100 items per page.
- W234712689 abstract "In this paper, we perform Chinese text categorization using n-gram text representation on TanCorpV1.0 which is a new corpus, special for Chinese text classification of more than 14,000 texts divided in 12 classes. We use a combination of methods, including between inter-class feature reduction methods and cross-class feature selection methods. We use the C-SVC classifier (with a linear kernel) which is the SVM algorithm made for the multi-classification task. We perform our experiments in the TANAGRA platform. Our experiments concern: (1) the performance comparison between using both 1-, 2-grams and using 1-, 2-, 3gram in Chinese text representation; (2) the performance comparison between using different feature representations: absolute text frequency, relative text frequency, absolute n-gram frequency and relative n-gram frequency; (3) the comparison of the sparseness in the “text*feature” matrix between using n-gram frequency and frequency in feature selection; (4) the performance comparison between two text coding methods: the 0/1 logical value and the n-gram frequency numeric value. We found out that in the case of using less than 3,000 features, the feature selection methods based on n-gram frequency (absolute or relative) always yield better results." @default.
- W234712689 created "2016-06-24" @default.
- W234712689 creator A5004301353 @default.
- W234712689 creator A5005477263 @default.
- W234712689 creator A5081986888 @default.
- W234712689 date "2008-01-01" @default.
- W234712689 modified "2023-09-27" @default.
- W234712689 title "Comparing different text representation and feature selection methods on Chinese text classification using Character n-grams" @default.
- W234712689 cites W1540550673 @default.
- W234712689 cites W2118020653 @default.
- W234712689 cites W2145345 @default.
- W234712689 cites W2149776783 @default.
- W234712689 cites W345244337 @default.
- W234712689 hasPublicationYear "2008" @default.
- W234712689 type Work @default.
- W234712689 sameAs 234712689 @default.
- W234712689 citedByCount "4" @default.
- W234712689 countsByYear W2347126892012 @default.
- W234712689 countsByYear W2347126892013 @default.
- W234712689 crossrefType "journal-article" @default.
- W234712689 hasAuthorship W234712689A5004301353 @default.
- W234712689 hasAuthorship W234712689A5005477263 @default.
- W234712689 hasAuthorship W234712689A5081986888 @default.
- W234712689 hasConcept C105795698 @default.
- W234712689 hasConcept C117884012 @default.
- W234712689 hasConcept C12267149 @default.
- W234712689 hasConcept C137293760 @default.
- W234712689 hasConcept C139532973 @default.
- W234712689 hasConcept C148483581 @default.
- W234712689 hasConcept C153180895 @default.
- W234712689 hasConcept C154945302 @default.
- W234712689 hasConcept C199075045 @default.
- W234712689 hasConcept C204321447 @default.
- W234712689 hasConcept C28490314 @default.
- W234712689 hasConcept C2986744138 @default.
- W234712689 hasConcept C33923547 @default.
- W234712689 hasConcept C41008148 @default.
- W234712689 hasConcept C95623464 @default.
- W234712689 hasConceptScore W234712689C105795698 @default.
- W234712689 hasConceptScore W234712689C117884012 @default.
- W234712689 hasConceptScore W234712689C12267149 @default.
- W234712689 hasConceptScore W234712689C137293760 @default.
- W234712689 hasConceptScore W234712689C139532973 @default.
- W234712689 hasConceptScore W234712689C148483581 @default.
- W234712689 hasConceptScore W234712689C153180895 @default.
- W234712689 hasConceptScore W234712689C154945302 @default.
- W234712689 hasConceptScore W234712689C199075045 @default.
- W234712689 hasConceptScore W234712689C204321447 @default.
- W234712689 hasConceptScore W234712689C28490314 @default.
- W234712689 hasConceptScore W234712689C2986744138 @default.
- W234712689 hasConceptScore W234712689C33923547 @default.
- W234712689 hasConceptScore W234712689C41008148 @default.
- W234712689 hasConceptScore W234712689C95623464 @default.
- W234712689 hasLocation W2347126891 @default.
- W234712689 hasOpenAccess W234712689 @default.
- W234712689 hasPrimaryLocation W2347126891 @default.
- W234712689 hasRelatedWork W1964438691 @default.
- W234712689 hasRelatedWork W1989962885 @default.
- W234712689 hasRelatedWork W2022689990 @default.
- W234712689 hasRelatedWork W2023384245 @default.
- W234712689 hasRelatedWork W2045041604 @default.
- W234712689 hasRelatedWork W2084868231 @default.
- W234712689 hasRelatedWork W2092285120 @default.
- W234712689 hasRelatedWork W2131912994 @default.
- W234712689 hasRelatedWork W2136114015 @default.
- W234712689 hasRelatedWork W2164008352 @default.
- W234712689 hasRelatedWork W2244814250 @default.
- W234712689 hasRelatedWork W2350640579 @default.
- W234712689 hasRelatedWork W2366349849 @default.
- W234712689 hasRelatedWork W2367691850 @default.
- W234712689 hasRelatedWork W2372055949 @default.
- W234712689 hasRelatedWork W2373990865 @default.
- W234712689 hasRelatedWork W2377478162 @default.
- W234712689 hasRelatedWork W2384119366 @default.
- W234712689 hasRelatedWork W2384422940 @default.
- W234712689 hasRelatedWork W2589247739 @default.
- W234712689 isParatext "false" @default.
- W234712689 isRetracted "false" @default.
- W234712689 magId "234712689" @default.
- W234712689 workType "article" @default.