Matches in SemOpenAlex for { <https://semopenalex.org/work/W3088816147> ?p ?o ?g. }
- W3088816147 abstract "In this paper, we focus on the imbalance issue, which is rarely studied in aspect term extraction and aspect sentiment classification when regarding them as sequence labeling tasks. Besides, previous works usually ignore the interaction between aspect terms when labeling polarities. We propose a GRadient hArmonized and CascadEd labeling model (GRACE) to solve these problems. Specifically, a cascaded labeling module is developed to enhance the interchange between aspect terms and improve the attention of sentiment tokens when labeling sentiment polarities. The polarities sequence is designed to depend on the generated aspect terms labels. To alleviate the imbalance issue, we extend the gradient harmonized mechanism used in object detection to the aspect-based sentiment analysis by adjusting the weight of each label dynamically. The proposed GRACE adopts a post-pretraining BERT as its backbone. Experimental results demonstrate that the proposed model achieves consistency improvement on multiple benchmark datasets and generates state-of-the-art results." @default.
- W3088816147 created "2020-10-01" @default.
- W3088816147 creator A5042018181 @default.
- W3088816147 creator A5060116992 @default.
- W3088816147 creator A5060295796 @default.
- W3088816147 creator A5070559820 @default.
- W3088816147 creator A5075965915 @default.
- W3088816147 date "2020-09-22" @default.
- W3088816147 modified "2023-09-27" @default.
- W3088816147 title "GRACE: Gradient Harmonized and Cascaded Labeling for Aspect-based Sentiment Analysis" @default.
- W3088816147 cites W2151752537 @default.
- W3088816147 cites W2159457224 @default.
- W3088816147 cites W2160660844 @default.
- W3088816147 cites W2234079371 @default.
- W3088816147 cites W2251294039 @default.
- W3088816147 cites W2251648804 @default.
- W3088816147 cites W2251900677 @default.
- W3088816147 cites W2252007242 @default.
- W3088816147 cites W2252012216 @default.
- W3088816147 cites W2427312199 @default.
- W3088816147 cites W2465978385 @default.
- W3088816147 cites W2556096924 @default.
- W3088816147 cites W2567277857 @default.
- W3088816147 cites W2604668619 @default.
- W3088816147 cites W2741252866 @default.
- W3088816147 cites W2799044502 @default.
- W3088816147 cites W2890240222 @default.
- W3088816147 cites W2925618549 @default.
- W3088816147 cites W2948947170 @default.
- W3088816147 cites W2949660355 @default.
- W3088816147 cites W2950404230 @default.
- W3088816147 cites W2950488390 @default.
- W3088816147 cites W2962741379 @default.
- W3088816147 cites W2962808042 @default.
- W3088816147 cites W2962843214 @default.
- W3088816147 cites W2963113370 @default.
- W3088816147 cites W2963264961 @default.
- W3088816147 cites W2963274454 @default.
- W3088816147 cites W2963341956 @default.
- W3088816147 cites W2963403868 @default.
- W3088816147 cites W2963428430 @default.
- W3088816147 cites W2963556938 @default.
- W3088816147 cites W2964059932 @default.
- W3088816147 cites W2964164368 @default.
- W3088816147 cites W2965510113 @default.
- W3088816147 cites W2997464781 @default.
- W3088816147 cites W2998446468 @default.
- W3088816147 cites W3003957883 @default.
- W3088816147 cites W3014324062 @default.
- W3088816147 cites W3014547480 @default.
- W3088816147 cites W3016552903 @default.
- W3088816147 cites W3034336785 @default.
- W3088816147 cites W3034990686 @default.
- W3088816147 cites W3035529900 @default.
- W3088816147 cites W3198311766 @default.
- W3088816147 doi "https://doi.org/10.48550/arxiv.2009.10557" @default.
- W3088816147 hasPublicationYear "2020" @default.
- W3088816147 type Work @default.
- W3088816147 sameAs 3088816147 @default.
- W3088816147 citedByCount "2" @default.
- W3088816147 countsByYear W30888161472020 @default.
- W3088816147 countsByYear W30888161472021 @default.
- W3088816147 crossrefType "posted-content" @default.
- W3088816147 hasAuthorship W3088816147A5042018181 @default.
- W3088816147 hasAuthorship W3088816147A5060116992 @default.
- W3088816147 hasAuthorship W3088816147A5060295796 @default.
- W3088816147 hasAuthorship W3088816147A5070559820 @default.
- W3088816147 hasAuthorship W3088816147A5075965915 @default.
- W3088816147 hasBestOaLocation W30888161471 @default.
- W3088816147 hasConcept C120665830 @default.
- W3088816147 hasConcept C121332964 @default.
- W3088816147 hasConcept C127413603 @default.
- W3088816147 hasConcept C13280743 @default.
- W3088816147 hasConcept C153180895 @default.
- W3088816147 hasConcept C154945302 @default.
- W3088816147 hasConcept C185592680 @default.
- W3088816147 hasConcept C185798385 @default.
- W3088816147 hasConcept C192209626 @default.
- W3088816147 hasConcept C201995342 @default.
- W3088816147 hasConcept C204321447 @default.
- W3088816147 hasConcept C205649164 @default.
- W3088816147 hasConcept C2776436953 @default.
- W3088816147 hasConcept C2778112365 @default.
- W3088816147 hasConcept C2780451532 @default.
- W3088816147 hasConcept C2781238097 @default.
- W3088816147 hasConcept C35639132 @default.
- W3088816147 hasConcept C41008148 @default.
- W3088816147 hasConcept C55493867 @default.
- W3088816147 hasConcept C66402592 @default.
- W3088816147 hasConceptScore W3088816147C120665830 @default.
- W3088816147 hasConceptScore W3088816147C121332964 @default.
- W3088816147 hasConceptScore W3088816147C127413603 @default.
- W3088816147 hasConceptScore W3088816147C13280743 @default.
- W3088816147 hasConceptScore W3088816147C153180895 @default.
- W3088816147 hasConceptScore W3088816147C154945302 @default.
- W3088816147 hasConceptScore W3088816147C185592680 @default.
- W3088816147 hasConceptScore W3088816147C185798385 @default.
- W3088816147 hasConceptScore W3088816147C192209626 @default.
- W3088816147 hasConceptScore W3088816147C201995342 @default.
- W3088816147 hasConceptScore W3088816147C204321447 @default.