Matches in SemOpenAlex for { <https://semopenalex.org/work/W4280497968> ?p ?o ?g. }
- W4280497968 abstract "In forestry studies, deep learning models have achieved excellent performance in many application scenarios (e.g., detecting forest damage). However, the unclear model decisions (i.e., black-box) undermine the credibility of the results and hinder their practicality. This study intends to obtain explanations of such models through the use of explainable artificial intelligence methods, and then use feature unlearning methods to improve their performance, which is the first such attempt in the field of forestry. Results of three experiments show that the model training can be guided by expertise to gain specific knowledge, which is reflected by explanations. For all three experiments based on synthetic and real leaf images, the improvement of models is quantified in the classification accuracy (up to 4.6%) and three indicators of explanation assessment (i.e., root-mean-square error, cosine similarity, and the proportion of important pixels). Besides, the introduced expertise in annotation matrix form was automatically created in all experiments. This study emphasizes that studies of deep learning in forestry should not only pursue model performance (e.g., higher classification accuracy) but also focus on the explanations and try to improve models according to the expertise." @default.
- W4280497968 created "2022-05-22" @default.
- W4280497968 creator A5012912213 @default.
- W4280497968 creator A5028731194 @default.
- W4280497968 creator A5087175500 @default.
- W4280497968 date "2022-05-19" @default.
- W4280497968 modified "2023-10-01" @default.
- W4280497968 title "Improve the Deep Learning Models in Forestry Based on Explanations and Expertise" @default.
- W4280497968 cites W1787224781 @default.
- W4280497968 cites W1977012749 @default.
- W4280497968 cites W2038541098 @default.
- W4280497968 cites W2124351162 @default.
- W4280497968 cites W2125847307 @default.
- W4280497968 cites W2469862374 @default.
- W4280497968 cites W2473156356 @default.
- W4280497968 cites W2528491735 @default.
- W4280497968 cites W2565950292 @default.
- W4280497968 cites W2618530766 @default.
- W4280497968 cites W2767814567 @default.
- W4280497968 cites W2782522152 @default.
- W4280497968 cites W2891130433 @default.
- W4280497968 cites W2921499963 @default.
- W4280497968 cites W2938959907 @default.
- W4280497968 cites W2958089299 @default.
- W4280497968 cites W2958514452 @default.
- W4280497968 cites W2963798744 @default.
- W4280497968 cites W2969634177 @default.
- W4280497968 cites W2974691105 @default.
- W4280497968 cites W3000716014 @default.
- W4280497968 cites W3031105282 @default.
- W4280497968 cites W3047930143 @default.
- W4280497968 cites W3048549109 @default.
- W4280497968 cites W3085610647 @default.
- W4280497968 cites W3093433874 @default.
- W4280497968 cites W3100711616 @default.
- W4280497968 cites W3120119699 @default.
- W4280497968 cites W3158308652 @default.
- W4280497968 cites W3173991935 @default.
- W4280497968 cites W3175558562 @default.
- W4280497968 cites W3184401950 @default.
- W4280497968 cites W3191346550 @default.
- W4280497968 cites W3057726865 @default.
- W4280497968 doi "https://doi.org/10.3389/fpls.2022.902105" @default.
- W4280497968 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/35677249" @default.
- W4280497968 hasPublicationYear "2022" @default.
- W4280497968 type Work @default.
- W4280497968 citedByCount "3" @default.
- W4280497968 countsByYear W42804979682022 @default.
- W4280497968 countsByYear W42804979682023 @default.
- W4280497968 crossrefType "journal-article" @default.
- W4280497968 hasAuthorship W4280497968A5012912213 @default.
- W4280497968 hasAuthorship W4280497968A5028731194 @default.
- W4280497968 hasAuthorship W4280497968A5087175500 @default.
- W4280497968 hasBestOaLocation W42804979681 @default.
- W4280497968 hasConcept C103278499 @default.
- W4280497968 hasConcept C108583219 @default.
- W4280497968 hasConcept C115961682 @default.
- W4280497968 hasConcept C119857082 @default.
- W4280497968 hasConcept C120665830 @default.
- W4280497968 hasConcept C121332964 @default.
- W4280497968 hasConcept C138885662 @default.
- W4280497968 hasConcept C154945302 @default.
- W4280497968 hasConcept C17744445 @default.
- W4280497968 hasConcept C192209626 @default.
- W4280497968 hasConcept C199539241 @default.
- W4280497968 hasConcept C202444582 @default.
- W4280497968 hasConcept C2776401178 @default.
- W4280497968 hasConcept C2780224610 @default.
- W4280497968 hasConcept C33923547 @default.
- W4280497968 hasConcept C41008148 @default.
- W4280497968 hasConcept C41895202 @default.
- W4280497968 hasConcept C9652623 @default.
- W4280497968 hasConceptScore W4280497968C103278499 @default.
- W4280497968 hasConceptScore W4280497968C108583219 @default.
- W4280497968 hasConceptScore W4280497968C115961682 @default.
- W4280497968 hasConceptScore W4280497968C119857082 @default.
- W4280497968 hasConceptScore W4280497968C120665830 @default.
- W4280497968 hasConceptScore W4280497968C121332964 @default.
- W4280497968 hasConceptScore W4280497968C138885662 @default.
- W4280497968 hasConceptScore W4280497968C154945302 @default.
- W4280497968 hasConceptScore W4280497968C17744445 @default.
- W4280497968 hasConceptScore W4280497968C192209626 @default.
- W4280497968 hasConceptScore W4280497968C199539241 @default.
- W4280497968 hasConceptScore W4280497968C202444582 @default.
- W4280497968 hasConceptScore W4280497968C2776401178 @default.
- W4280497968 hasConceptScore W4280497968C2780224610 @default.
- W4280497968 hasConceptScore W4280497968C33923547 @default.
- W4280497968 hasConceptScore W4280497968C41008148 @default.
- W4280497968 hasConceptScore W4280497968C41895202 @default.
- W4280497968 hasConceptScore W4280497968C9652623 @default.
- W4280497968 hasFunder F4320321114 @default.
- W4280497968 hasLocation W42804979681 @default.
- W4280497968 hasLocation W42804979682 @default.
- W4280497968 hasLocation W42804979683 @default.
- W4280497968 hasOpenAccess W4280497968 @default.
- W4280497968 hasPrimaryLocation W42804979681 @default.
- W4280497968 hasRelatedWork W2795261237 @default.
- W4280497968 hasRelatedWork W3014300295 @default.
- W4280497968 hasRelatedWork W3164822677 @default.
- W4280497968 hasRelatedWork W4223943233 @default.