Matches in SemOpenAlex for { <https://semopenalex.org/work/W4294579467> ?p ?o ?g. }
- W4294579467 endingPage "347" @default.
- W4294579467 startingPage "340" @default.
- W4294579467 abstract "Extractive document summarization is a fundamental task in natural language processing (NLP). Recently, several Graph Neural Networks (GNNs) are proposed for this task. However, most existing GNN-based models can neither effectively encode semantic nodes of multiple granularity level apart from sentences nor substantially capture different cross-sentence meta-paths. To address these issues, we propose MHgatSum, a novel Multi-granularity Heterogeneous Graph ATtention networks for extractive document SUMmarization. Specifically, we first build a multi-granularity heterogeneous graph (HetG) for each document, which is better to represent the semantic meaning of the document. The HetG contains not only sentence nodes but also multiple other granularity effective semantic units with different semantic levels, including keyphrases and topics. These additional nodes act as the intermediary between sentences to build the meta-paths involved in sentence node (i.e., Sentence-Keyphrase-Sentence and Sentence-Topic-Sentence). Then, we propose a heterogeneous graph attention networks to embed the constructed HetG for extractive summarization, which enjoys multi-granularity semantic representations. The model is based on a hierarchical attention mechanism, including node-level and semantic-level attentions. The node-level attention can learn the importance between a node and its meta-path based neighbors, while the semantic-level attention is able to learn the importance of different meta-paths. Moreover, to better integrate sentence global knowledge, we further incorporate sentence node global importance in local node-level attention. We conduct empirical experiments on two benchmark datasets, which demonstrates the superiority of MHgatSum over previous SOTA models on the task of extractive summarization." @default.
- W4294579467 created "2022-09-05" @default.
- W4294579467 creator A5004325176 @default.
- W4294579467 creator A5029854281 @default.
- W4294579467 creator A5039067585 @default.
- W4294579467 creator A5051167590 @default.
- W4294579467 creator A5051534020 @default.
- W4294579467 creator A5053780153 @default.
- W4294579467 creator A5056219928 @default.
- W4294579467 creator A5081246710 @default.
- W4294579467 date "2022-11-01" @default.
- W4294579467 modified "2023-09-23" @default.
- W4294579467 title "Multi-granularity heterogeneous graph attention networks for extractive document summarization" @default.
- W4294579467 cites W2150824314 @default.
- W4294579467 cites W2307381258 @default.
- W4294579467 cites W2606974598 @default.
- W4294579467 cites W2890419630 @default.
- W4294579467 cites W2892094955 @default.
- W4294579467 cites W2896807716 @default.
- W4294579467 cites W2911286998 @default.
- W4294579467 cites W2950342809 @default.
- W4294579467 cites W2951682790 @default.
- W4294579467 cites W2952138241 @default.
- W4294579467 cites W2962785754 @default.
- W4294579467 cites W2962946486 @default.
- W4294579467 cites W2962965405 @default.
- W4294579467 cites W2962972512 @default.
- W4294579467 cites W2963245897 @default.
- W4294579467 cites W2963251530 @default.
- W4294579467 cites W2963385935 @default.
- W4294579467 cites W2963964898 @default.
- W4294579467 cites W2963968665 @default.
- W4294579467 cites W2964028111 @default.
- W4294579467 cites W2964144561 @default.
- W4294579467 cites W2970419734 @default.
- W4294579467 cites W2970830889 @default.
- W4294579467 cites W2971289520 @default.
- W4294579467 cites W2985619053 @default.
- W4294579467 cites W3034353423 @default.
- W4294579467 cites W3034682120 @default.
- W4294579467 cites W3034961030 @default.
- W4294579467 cites W3035050380 @default.
- W4294579467 cites W3035643691 @default.
- W4294579467 cites W3100053428 @default.
- W4294579467 cites W3116498179 @default.
- W4294579467 cites W3117801652 @default.
- W4294579467 cites W3138773240 @default.
- W4294579467 cites W2930279335 @default.
- W4294579467 doi "https://doi.org/10.1016/j.neunet.2022.08.021" @default.
- W4294579467 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/36113341" @default.
- W4294579467 hasPublicationYear "2022" @default.
- W4294579467 type Work @default.
- W4294579467 citedByCount "3" @default.
- W4294579467 countsByYear W42945794672022 @default.
- W4294579467 countsByYear W42945794672023 @default.
- W4294579467 crossrefType "journal-article" @default.
- W4294579467 hasAuthorship W4294579467A5004325176 @default.
- W4294579467 hasAuthorship W4294579467A5029854281 @default.
- W4294579467 hasAuthorship W4294579467A5039067585 @default.
- W4294579467 hasAuthorship W4294579467A5051167590 @default.
- W4294579467 hasAuthorship W4294579467A5051534020 @default.
- W4294579467 hasAuthorship W4294579467A5053780153 @default.
- W4294579467 hasAuthorship W4294579467A5056219928 @default.
- W4294579467 hasAuthorship W4294579467A5081246710 @default.
- W4294579467 hasConcept C111919701 @default.
- W4294579467 hasConcept C127413603 @default.
- W4294579467 hasConcept C132525143 @default.
- W4294579467 hasConcept C154945302 @default.
- W4294579467 hasConcept C170858558 @default.
- W4294579467 hasConcept C177774035 @default.
- W4294579467 hasConcept C204321447 @default.
- W4294579467 hasConcept C2777530160 @default.
- W4294579467 hasConcept C41008148 @default.
- W4294579467 hasConcept C62611344 @default.
- W4294579467 hasConcept C66938386 @default.
- W4294579467 hasConcept C80444323 @default.
- W4294579467 hasConceptScore W4294579467C111919701 @default.
- W4294579467 hasConceptScore W4294579467C127413603 @default.
- W4294579467 hasConceptScore W4294579467C132525143 @default.
- W4294579467 hasConceptScore W4294579467C154945302 @default.
- W4294579467 hasConceptScore W4294579467C170858558 @default.
- W4294579467 hasConceptScore W4294579467C177774035 @default.
- W4294579467 hasConceptScore W4294579467C204321447 @default.
- W4294579467 hasConceptScore W4294579467C2777530160 @default.
- W4294579467 hasConceptScore W4294579467C41008148 @default.
- W4294579467 hasConceptScore W4294579467C62611344 @default.
- W4294579467 hasConceptScore W4294579467C66938386 @default.
- W4294579467 hasConceptScore W4294579467C80444323 @default.
- W4294579467 hasLocation W42945794671 @default.
- W4294579467 hasLocation W42945794672 @default.
- W4294579467 hasOpenAccess W4294579467 @default.
- W4294579467 hasPrimaryLocation W42945794671 @default.
- W4294579467 hasRelatedWork W2035950535 @default.
- W4294579467 hasRelatedWork W2104752822 @default.
- W4294579467 hasRelatedWork W2293457016 @default.
- W4294579467 hasRelatedWork W2330186386 @default.
- W4294579467 hasRelatedWork W2347941600 @default.
- W4294579467 hasRelatedWork W2401226416 @default.