Matches in SemOpenAlex for { <https://semopenalex.org/work/W3173664641> ?p ?o ?g. }
- W3173664641 endingPage "102659" @default.
- W3173664641 startingPage "102659" @default.
- W3173664641 abstract "Fact verification aims to retrieve related evidence from raw text to verify the correctness of a given claim. Existing works mainly leverage the single-granularity features for the representation learning of evidences, i.e., sentence features, ignoring other features like entity-level and context-level features. In addition, they usually focus on improving the prediction accuracy while lacking the interpretability of the inference process, which leads to unreliable results. Thus, in this paper, to investigate how to utilize multi-granularity semantic units for evidence representation as well as to improve the explainability of evidence reasoning, we propose a H ierarchical Reasoning-based H eterogeneous G raph Neural N etwork for fact verification (HHGN). HHGN combines multiple features of entity, sentence as well as context for evidence representation, and employs a heterogeneous graph to capture their semantic relations. Inspired by the human inference process, we design a hierarchical reasoning-based node updating strategy to propagate the evidence features. Then, we extract the potential reasoning paths from the graph to predict the label, which aggregates the results of different paths weighted by their relevance to the claim. We evaluate our proposal on FEVER, a large-scale benchmark dataset for fact verification. Our experimental results demonstrate the superiority of HHGN over the competitive baselines in both single evidence and multiple evidences settings. In addition, HHGN presents reasonable interpretability in the form of aggregating the features of relevant entity units and selecting the evidence sentences with high confidence. • A heterogeneous graph-based evidence representation method is proposed. • A hierarchical reasoning-based node features aggregation strategy is designed. • Experimental results demonstrate the superiority of our proposal on both multiple and single evidence scenarios." @default.
- W3173664641 created "2021-07-05" @default.
- W3173664641 creator A5026327666 @default.
- W3173664641 creator A5028271019 @default.
- W3173664641 creator A5031127943 @default.
- W3173664641 creator A5054702314 @default.
- W3173664641 creator A5066699855 @default.
- W3173664641 date "2021-09-01" @default.
- W3173664641 modified "2023-10-16" @default.
- W3173664641 title "HHGN: A Hierarchical Reasoning-based Heterogeneous Graph Neural Network for fact verification" @default.
- W3173664641 cites W2116341502 @default.
- W3173664641 cites W2163922914 @default.
- W3173664641 cites W2250539671 @default.
- W3173664641 cites W2507974895 @default.
- W3173664641 cites W2608787653 @default.
- W3173664641 cites W2788667846 @default.
- W3173664641 cites W2802187397 @default.
- W3173664641 cites W2886390638 @default.
- W3173664641 cites W2892163801 @default.
- W3173664641 cites W2950031296 @default.
- W3173664641 cites W2950336186 @default.
- W3173664641 cites W2951682790 @default.
- W3173664641 cites W2951862794 @default.
- W3173664641 cites W2955452806 @default.
- W3173664641 cites W2955492726 @default.
- W3173664641 cites W2955753753 @default.
- W3173664641 cites W2962946486 @default.
- W3173664641 cites W2962985038 @default.
- W3173664641 cites W2963691697 @default.
- W3173664641 cites W2963961878 @default.
- W3173664641 cites W2963964898 @default.
- W3173664641 cites W2964060837 @default.
- W3173664641 cites W2964068236 @default.
- W3173664641 cites W2970139579 @default.
- W3173664641 cites W2987294674 @default.
- W3173664641 cites W3017344694 @default.
- W3173664641 cites W3034353423 @default.
- W3173664641 cites W3034808961 @default.
- W3173664641 cites W3034828027 @default.
- W3173664641 cites W3038519182 @default.
- W3173664641 cites W3058800094 @default.
- W3173664641 cites W3100790518 @default.
- W3173664641 cites W3118280924 @default.
- W3173664641 doi "https://doi.org/10.1016/j.ipm.2021.102659" @default.
- W3173664641 hasPublicationYear "2021" @default.
- W3173664641 type Work @default.
- W3173664641 sameAs 3173664641 @default.
- W3173664641 citedByCount "10" @default.
- W3173664641 countsByYear W31736646412021 @default.
- W3173664641 countsByYear W31736646412022 @default.
- W3173664641 countsByYear W31736646412023 @default.
- W3173664641 crossrefType "journal-article" @default.
- W3173664641 hasAuthorship W3173664641A5026327666 @default.
- W3173664641 hasAuthorship W3173664641A5028271019 @default.
- W3173664641 hasAuthorship W3173664641A5031127943 @default.
- W3173664641 hasAuthorship W3173664641A5054702314 @default.
- W3173664641 hasAuthorship W3173664641A5066699855 @default.
- W3173664641 hasConcept C119857082 @default.
- W3173664641 hasConcept C132525143 @default.
- W3173664641 hasConcept C153083717 @default.
- W3173664641 hasConcept C154945302 @default.
- W3173664641 hasConcept C199360897 @default.
- W3173664641 hasConcept C204321447 @default.
- W3173664641 hasConcept C2776214188 @default.
- W3173664641 hasConcept C2777530160 @default.
- W3173664641 hasConcept C2781067378 @default.
- W3173664641 hasConcept C41008148 @default.
- W3173664641 hasConcept C55439883 @default.
- W3173664641 hasConcept C80444323 @default.
- W3173664641 hasConceptScore W3173664641C119857082 @default.
- W3173664641 hasConceptScore W3173664641C132525143 @default.
- W3173664641 hasConceptScore W3173664641C153083717 @default.
- W3173664641 hasConceptScore W3173664641C154945302 @default.
- W3173664641 hasConceptScore W3173664641C199360897 @default.
- W3173664641 hasConceptScore W3173664641C204321447 @default.
- W3173664641 hasConceptScore W3173664641C2776214188 @default.
- W3173664641 hasConceptScore W3173664641C2777530160 @default.
- W3173664641 hasConceptScore W3173664641C2781067378 @default.
- W3173664641 hasConceptScore W3173664641C41008148 @default.
- W3173664641 hasConceptScore W3173664641C55439883 @default.
- W3173664641 hasConceptScore W3173664641C80444323 @default.
- W3173664641 hasFunder F4320321001 @default.
- W3173664641 hasFunder F4320326217 @default.
- W3173664641 hasIssue "5" @default.
- W3173664641 hasLocation W31736646411 @default.
- W3173664641 hasOpenAccess W3173664641 @default.
- W3173664641 hasPrimaryLocation W31736646411 @default.
- W3173664641 hasRelatedWork W1517743118 @default.
- W3173664641 hasRelatedWork W3006943036 @default.
- W3173664641 hasRelatedWork W3012234327 @default.
- W3173664641 hasRelatedWork W3107474891 @default.
- W3173664641 hasRelatedWork W3191046242 @default.
- W3173664641 hasRelatedWork W3203961807 @default.
- W3173664641 hasRelatedWork W4205364923 @default.
- W3173664641 hasRelatedWork W4206534706 @default.
- W3173664641 hasRelatedWork W4229079080 @default.
- W3173664641 hasRelatedWork W4294031299 @default.
- W3173664641 hasVolume "58" @default.