Matches in SemOpenAlex for { <https://semopenalex.org/work/W3017250206> ?p ?o ?g. }
Showing items 1 to 60 of
60
with 100 items per page.
- W3017250206 abstract "Conventional methods of medical text data classification, neglect of context among different words and semantic information, has a poor text description, classification effect and generalization capability and robustness. To tackle the inefficiencies and low precision in the classification of medical text data, in this paper, we presented a new classification method with improved convolutional neural network (CNN) and support vector machine (SVM), i.e., CNN-SVM method. In the method, some convolution kernel filters that contribute greatly to the CNN model are first selected by the average response energy (ARE) value, and then used to simplify and reconstruct the CNN model. Next, the SVM classifier was optimized by firefly algorithm (FA) and context information to overcome the disadvantages of over-saturation and over-training in SVM classification. Finally, the presented CNN-SVM method is tested by the simulation experiment and the true classification of medical text data. The experimental results show that the presented CNN-SVM method in this paper can significantly reduce the complexity and amount of computation compared to the conventional methods, and further promote the computational efficiency and classification accuracy of medical text data." @default.
- W3017250206 created "2020-04-24" @default.
- W3017250206 creator A5035128762 @default.
- W3017250206 creator A5037457403 @default.
- W3017250206 creator A5072860776 @default.
- W3017250206 creator A5090035031 @default.
- W3017250206 date "2020-07-01" @default.
- W3017250206 modified "2023-09-25" @default.
- W3017250206 title "Classification of Medical Text Data Using Convolutional Neural Network-Support Vector Machine Method" @default.
- W3017250206 doi "https://doi.org/10.1166/jmihi.2020.3042" @default.
- W3017250206 hasPublicationYear "2020" @default.
- W3017250206 type Work @default.
- W3017250206 sameAs 3017250206 @default.
- W3017250206 citedByCount "3" @default.
- W3017250206 countsByYear W30172502062021 @default.
- W3017250206 countsByYear W30172502062022 @default.
- W3017250206 crossrefType "journal-article" @default.
- W3017250206 hasAuthorship W3017250206A5035128762 @default.
- W3017250206 hasAuthorship W3017250206A5037457403 @default.
- W3017250206 hasAuthorship W3017250206A5072860776 @default.
- W3017250206 hasAuthorship W3017250206A5090035031 @default.
- W3017250206 hasConcept C104317684 @default.
- W3017250206 hasConcept C119857082 @default.
- W3017250206 hasConcept C12267149 @default.
- W3017250206 hasConcept C153180895 @default.
- W3017250206 hasConcept C154945302 @default.
- W3017250206 hasConcept C185592680 @default.
- W3017250206 hasConcept C41008148 @default.
- W3017250206 hasConcept C55493867 @default.
- W3017250206 hasConcept C63479239 @default.
- W3017250206 hasConcept C81363708 @default.
- W3017250206 hasConcept C95623464 @default.
- W3017250206 hasConceptScore W3017250206C104317684 @default.
- W3017250206 hasConceptScore W3017250206C119857082 @default.
- W3017250206 hasConceptScore W3017250206C12267149 @default.
- W3017250206 hasConceptScore W3017250206C153180895 @default.
- W3017250206 hasConceptScore W3017250206C154945302 @default.
- W3017250206 hasConceptScore W3017250206C185592680 @default.
- W3017250206 hasConceptScore W3017250206C41008148 @default.
- W3017250206 hasConceptScore W3017250206C55493867 @default.
- W3017250206 hasConceptScore W3017250206C63479239 @default.
- W3017250206 hasConceptScore W3017250206C81363708 @default.
- W3017250206 hasConceptScore W3017250206C95623464 @default.
- W3017250206 hasLocation W30172502061 @default.
- W3017250206 hasOpenAccess W3017250206 @default.
- W3017250206 hasPrimaryLocation W30172502061 @default.
- W3017250206 hasRelatedWork W1224884 @default.
- W3017250206 hasRelatedWork W12783365 @default.
- W3017250206 hasRelatedWork W13034104 @default.
- W3017250206 hasRelatedWork W1368183 @default.
- W3017250206 hasRelatedWork W14328740 @default.
- W3017250206 hasRelatedWork W4703449 @default.
- W3017250206 hasRelatedWork W6229082 @default.
- W3017250206 hasRelatedWork W6717794 @default.
- W3017250206 hasRelatedWork W8220366 @default.
- W3017250206 hasRelatedWork W9778490 @default.
- W3017250206 isParatext "false" @default.
- W3017250206 isRetracted "false" @default.
- W3017250206 magId "3017250206" @default.
- W3017250206 workType "article" @default.