Matches in SemOpenAlex for { <https://semopenalex.org/work/W3199474611> ?p ?o ?g. }
Showing items 1 to 90 of
90
with 100 items per page.
- W3199474611 abstract "The normalized cross-correlation coefficient is a standard metric that can be used to evaluate the relation of one vector with another. In machine learning and as part of feature selection, the coefficient can be used to evaluate the importance of each feature in relation to the required output. This work aims to utilize this coefficient or similar feature importance indicators to improve the accuracy of a neural network by first computing the corresponding coefficient of each of the input features with the required output and then by creating a specialized network layer that can be used to improve the accuracy of the network. The specialized and novel layer requires adding a number of parameters that depends on the number of input features to the network. These parameters are optimized through the learning algorithm utilized by the network. The proposed layer guides the network to perform better with little effect on the computational cost by initially weighing each feature according to its importance such as the correlation. It is possible to use meaningful coefficients such as the Pearson correlation, the Spearman coefficient or others. The results indicate a consistently enhanced performance with the accuracy gain often exceeding 1% to more than 5% in some cases. The work considers several types of data including regression datasets, multi-output and convolutional neural networks related problems with concentration on binary classification datasets." @default.
- W3199474611 created "2021-09-27" @default.
- W3199474611 creator A5067321490 @default.
- W3199474611 creator A5075443955 @default.
- W3199474611 date "2021-07-18" @default.
- W3199474611 modified "2023-09-27" @default.
- W3199474611 title "A Novel Feature Importance Based Layer to Improve Neural Networks" @default.
- W3199474611 cites W1529026127 @default.
- W3199474611 cites W2027666102 @default.
- W3199474611 cites W2050308220 @default.
- W3199474611 cites W2064675550 @default.
- W3199474611 cites W2080218231 @default.
- W3199474611 cites W2132862423 @default.
- W3199474611 cites W2157331557 @default.
- W3199474611 cites W3099878876 @default.
- W3199474611 cites W4236706032 @default.
- W3199474611 doi "https://doi.org/10.1109/ijcnn52387.2021.9533691" @default.
- W3199474611 hasPublicationYear "2021" @default.
- W3199474611 type Work @default.
- W3199474611 sameAs 3199474611 @default.
- W3199474611 citedByCount "1" @default.
- W3199474611 countsByYear W31994746112022 @default.
- W3199474611 crossrefType "proceedings-article" @default.
- W3199474611 hasAuthorship W3199474611A5067321490 @default.
- W3199474611 hasAuthorship W3199474611A5075443955 @default.
- W3199474611 hasConcept C105795698 @default.
- W3199474611 hasConcept C117220453 @default.
- W3199474611 hasConcept C119857082 @default.
- W3199474611 hasConcept C124101348 @default.
- W3199474611 hasConcept C127413603 @default.
- W3199474611 hasConcept C138885662 @default.
- W3199474611 hasConcept C148483581 @default.
- W3199474611 hasConcept C153180895 @default.
- W3199474611 hasConcept C154945302 @default.
- W3199474611 hasConcept C176217482 @default.
- W3199474611 hasConcept C178790620 @default.
- W3199474611 hasConcept C185592680 @default.
- W3199474611 hasConcept C21547014 @default.
- W3199474611 hasConcept C2524010 @default.
- W3199474611 hasConcept C25343380 @default.
- W3199474611 hasConcept C2776401178 @default.
- W3199474611 hasConcept C2779227376 @default.
- W3199474611 hasConcept C2780092901 @default.
- W3199474611 hasConcept C33923547 @default.
- W3199474611 hasConcept C41008148 @default.
- W3199474611 hasConcept C41895202 @default.
- W3199474611 hasConcept C50644808 @default.
- W3199474611 hasConcept C55078378 @default.
- W3199474611 hasConcept C81363708 @default.
- W3199474611 hasConceptScore W3199474611C105795698 @default.
- W3199474611 hasConceptScore W3199474611C117220453 @default.
- W3199474611 hasConceptScore W3199474611C119857082 @default.
- W3199474611 hasConceptScore W3199474611C124101348 @default.
- W3199474611 hasConceptScore W3199474611C127413603 @default.
- W3199474611 hasConceptScore W3199474611C138885662 @default.
- W3199474611 hasConceptScore W3199474611C148483581 @default.
- W3199474611 hasConceptScore W3199474611C153180895 @default.
- W3199474611 hasConceptScore W3199474611C154945302 @default.
- W3199474611 hasConceptScore W3199474611C176217482 @default.
- W3199474611 hasConceptScore W3199474611C178790620 @default.
- W3199474611 hasConceptScore W3199474611C185592680 @default.
- W3199474611 hasConceptScore W3199474611C21547014 @default.
- W3199474611 hasConceptScore W3199474611C2524010 @default.
- W3199474611 hasConceptScore W3199474611C25343380 @default.
- W3199474611 hasConceptScore W3199474611C2776401178 @default.
- W3199474611 hasConceptScore W3199474611C2779227376 @default.
- W3199474611 hasConceptScore W3199474611C2780092901 @default.
- W3199474611 hasConceptScore W3199474611C33923547 @default.
- W3199474611 hasConceptScore W3199474611C41008148 @default.
- W3199474611 hasConceptScore W3199474611C41895202 @default.
- W3199474611 hasConceptScore W3199474611C50644808 @default.
- W3199474611 hasConceptScore W3199474611C55078378 @default.
- W3199474611 hasConceptScore W3199474611C81363708 @default.
- W3199474611 hasLocation W31994746111 @default.
- W3199474611 hasOpenAccess W3199474611 @default.
- W3199474611 hasPrimaryLocation W31994746111 @default.
- W3199474611 hasRelatedWork W157430051 @default.
- W3199474611 hasRelatedWork W2358421605 @default.
- W3199474611 hasRelatedWork W2760085659 @default.
- W3199474611 hasRelatedWork W2891907751 @default.
- W3199474611 hasRelatedWork W2903737112 @default.
- W3199474611 hasRelatedWork W2999681686 @default.
- W3199474611 hasRelatedWork W4206968802 @default.
- W3199474611 hasRelatedWork W4229019954 @default.
- W3199474611 hasRelatedWork W4307883119 @default.
- W3199474611 hasRelatedWork W2114089515 @default.
- W3199474611 isParatext "false" @default.
- W3199474611 isRetracted "false" @default.
- W3199474611 magId "3199474611" @default.
- W3199474611 workType "article" @default.