Matches in SemOpenAlex for { <https://semopenalex.org/work/W2788893468> ?p ?o ?g. }
Showing items 1 to 71 of
71
with 100 items per page.
- W2788893468 abstract "We mainly investigate word influence in neural sentiment classification, which results in a novel approach to promoting word sentiment and negation as attentions. Particularly, a sentiment and negation neural network (SNNN) is proposed, including a sentiment neural network (SNN) and a negation neural network (NNN). First, we modify the word level by embedding the word sentiment and negation information as the extra layers for the input. Second, we adopt a hierarchical LSTM model to generate the word-level, sentence-level and document-level representations respectively. After that, we enhance word sentiment and negation as attentions over the semantic level. Finally, the experiments conducting on the IMDB and Yelp data sets show that our approach is superior to the state-of-the-art baselines. Furthermore, we draw the interesting conclusions that (1) LSTM performs better than CNN and RNN for neural sentiment classification; (2) word sentiment and negation are a strong alliance with attention, while overfitting occurs when they are simultaneously applied at the embedding layer; and (3) word sentiment/negation can be singly implemented for better performance as both embedding layer and attention at the same time." @default.
- W2788893468 created "2018-03-06" @default.
- W2788893468 creator A5007834164 @default.
- W2788893468 creator A5018749138 @default.
- W2788893468 creator A5071451609 @default.
- W2788893468 creator A5083964406 @default.
- W2788893468 date "2018-04-29" @default.
- W2788893468 modified "2023-10-16" @default.
- W2788893468 title "SNNN: Promoting Word Sentiment and Negation in Neural Sentiment Classification" @default.
- W2788893468 doi "https://doi.org/10.1609/aaai.v32i1.11665" @default.
- W2788893468 hasPublicationYear "2018" @default.
- W2788893468 type Work @default.
- W2788893468 sameAs 2788893468 @default.
- W2788893468 citedByCount "5" @default.
- W2788893468 countsByYear W27888934682019 @default.
- W2788893468 countsByYear W27888934682020 @default.
- W2788893468 countsByYear W27888934682021 @default.
- W2788893468 countsByYear W27888934682022 @default.
- W2788893468 crossrefType "journal-article" @default.
- W2788893468 hasAuthorship W2788893468A5007834164 @default.
- W2788893468 hasAuthorship W2788893468A5018749138 @default.
- W2788893468 hasAuthorship W2788893468A5071451609 @default.
- W2788893468 hasAuthorship W2788893468A5083964406 @default.
- W2788893468 hasBestOaLocation W27888934681 @default.
- W2788893468 hasConcept C138885662 @default.
- W2788893468 hasConcept C154945302 @default.
- W2788893468 hasConcept C199360897 @default.
- W2788893468 hasConcept C204321447 @default.
- W2788893468 hasConcept C2185349 @default.
- W2788893468 hasConcept C22019652 @default.
- W2788893468 hasConcept C2777462759 @default.
- W2788893468 hasConcept C2777530160 @default.
- W2788893468 hasConcept C41008148 @default.
- W2788893468 hasConcept C41608201 @default.
- W2788893468 hasConcept C41895202 @default.
- W2788893468 hasConcept C50644808 @default.
- W2788893468 hasConcept C66402592 @default.
- W2788893468 hasConcept C90805587 @default.
- W2788893468 hasConceptScore W2788893468C138885662 @default.
- W2788893468 hasConceptScore W2788893468C154945302 @default.
- W2788893468 hasConceptScore W2788893468C199360897 @default.
- W2788893468 hasConceptScore W2788893468C204321447 @default.
- W2788893468 hasConceptScore W2788893468C2185349 @default.
- W2788893468 hasConceptScore W2788893468C22019652 @default.
- W2788893468 hasConceptScore W2788893468C2777462759 @default.
- W2788893468 hasConceptScore W2788893468C2777530160 @default.
- W2788893468 hasConceptScore W2788893468C41008148 @default.
- W2788893468 hasConceptScore W2788893468C41608201 @default.
- W2788893468 hasConceptScore W2788893468C41895202 @default.
- W2788893468 hasConceptScore W2788893468C50644808 @default.
- W2788893468 hasConceptScore W2788893468C66402592 @default.
- W2788893468 hasConceptScore W2788893468C90805587 @default.
- W2788893468 hasIssue "1" @default.
- W2788893468 hasLocation W27888934681 @default.
- W2788893468 hasOpenAccess W2788893468 @default.
- W2788893468 hasPrimaryLocation W27888934681 @default.
- W2788893468 hasRelatedWork W2140548485 @default.
- W2788893468 hasRelatedWork W2335882425 @default.
- W2788893468 hasRelatedWork W2338093180 @default.
- W2788893468 hasRelatedWork W2543875770 @default.
- W2788893468 hasRelatedWork W2745862583 @default.
- W2788893468 hasRelatedWork W2788893468 @default.
- W2788893468 hasRelatedWork W3107679445 @default.
- W2788893468 hasRelatedWork W3147405789 @default.
- W2788893468 hasRelatedWork W4200241356 @default.
- W2788893468 hasRelatedWork W4323981018 @default.
- W2788893468 hasVolume "32" @default.
- W2788893468 isParatext "false" @default.
- W2788893468 isRetracted "false" @default.
- W2788893468 magId "2788893468" @default.
- W2788893468 workType "article" @default.