Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386004948> ?p ?o ?g. }
Showing items 1 to 79 of
79
with 100 items per page.
- W4386004948 endingPage "39" @default.
- W4386004948 startingPage "25" @default.
- W4386004948 abstract "Business documents are used every day by all kinds and sizes of companies and administrations, even if most of these entities have several information systems where the documents are digitilized in different formats (json, xml, database tables, ...), there still an important number of business documents that require manual processing which costs a lot and is very time consuming. Extracting key-value information from business documents is a challenging problem due to the variety of document types and templates, in this work we will deal with the problem as a graph node classification problem using a multi “graph transformer layers, we propose a graph construction method that focuses on the most relevant neighbours of every node while reducing the size of the graph and we use a document transformer embedding combined with some spatial and textual feature to give a better representation to each node. Our contribution in this work was to conceive a graph neural network (GNN) achieving the highest results comparing to the rest of GNN models dealing with the same problem to our knowledge, the model is small (53,6K parameters) comparing to the recent models using transformers architectures (hundreds of millions of parameters) which is very suitable for applications when storage constraints are present, it also has a limited impact on the environment and represent an alternative to build greener AI systems. We experiment our model on the SROIE ICDAR receipts dataset where we got an important F1 score compared to other graph neural network (GNN) based models." @default.
- W4386004948 created "2023-08-20" @default.
- W4386004948 creator A5032791589 @default.
- W4386004948 creator A5040771608 @default.
- W4386004948 creator A5042793065 @default.
- W4386004948 creator A5067446887 @default.
- W4386004948 date "2023-01-01" @default.
- W4386004948 modified "2023-10-12" @default.
- W4386004948 title "Enhancing GNN Feature Modeling for Document Information Extraction Using Transformers" @default.
- W4386004948 cites W2059192331 @default.
- W4386004948 cites W2112796928 @default.
- W4386004948 cites W2135164809 @default.
- W4386004948 cites W2891117443 @default.
- W4386004948 cites W2922714365 @default.
- W4386004948 cites W2951087161 @default.
- W4386004948 cites W2962772269 @default.
- W4386004948 cites W2998913931 @default.
- W4386004948 cites W3003261556 @default.
- W4386004948 cites W3003484198 @default.
- W4386004948 cites W3004330198 @default.
- W4386004948 cites W3034864438 @default.
- W4386004948 cites W3104953317 @default.
- W4386004948 cites W3113753692 @default.
- W4386004948 cites W3187966659 @default.
- W4386004948 cites W3196589123 @default.
- W4386004948 cites W3205981739 @default.
- W4386004948 doi "https://doi.org/10.1007/978-3-031-40773-4_2" @default.
- W4386004948 hasPublicationYear "2023" @default.
- W4386004948 type Work @default.
- W4386004948 citedByCount "0" @default.
- W4386004948 crossrefType "book-chapter" @default.
- W4386004948 hasAuthorship W4386004948A5032791589 @default.
- W4386004948 hasAuthorship W4386004948A5040771608 @default.
- W4386004948 hasAuthorship W4386004948A5042793065 @default.
- W4386004948 hasAuthorship W4386004948A5067446887 @default.
- W4386004948 hasConcept C121332964 @default.
- W4386004948 hasConcept C124101348 @default.
- W4386004948 hasConcept C132525143 @default.
- W4386004948 hasConcept C136764020 @default.
- W4386004948 hasConcept C154945302 @default.
- W4386004948 hasConcept C165801399 @default.
- W4386004948 hasConcept C23123220 @default.
- W4386004948 hasConcept C41008148 @default.
- W4386004948 hasConcept C41608201 @default.
- W4386004948 hasConcept C62520636 @default.
- W4386004948 hasConcept C66322947 @default.
- W4386004948 hasConcept C80444323 @default.
- W4386004948 hasConcept C8797682 @default.
- W4386004948 hasConceptScore W4386004948C121332964 @default.
- W4386004948 hasConceptScore W4386004948C124101348 @default.
- W4386004948 hasConceptScore W4386004948C132525143 @default.
- W4386004948 hasConceptScore W4386004948C136764020 @default.
- W4386004948 hasConceptScore W4386004948C154945302 @default.
- W4386004948 hasConceptScore W4386004948C165801399 @default.
- W4386004948 hasConceptScore W4386004948C23123220 @default.
- W4386004948 hasConceptScore W4386004948C41008148 @default.
- W4386004948 hasConceptScore W4386004948C41608201 @default.
- W4386004948 hasConceptScore W4386004948C62520636 @default.
- W4386004948 hasConceptScore W4386004948C66322947 @default.
- W4386004948 hasConceptScore W4386004948C80444323 @default.
- W4386004948 hasConceptScore W4386004948C8797682 @default.
- W4386004948 hasLocation W43860049481 @default.
- W4386004948 hasOpenAccess W4386004948 @default.
- W4386004948 hasPrimaryLocation W43860049481 @default.
- W4386004948 hasRelatedWork W2088247287 @default.
- W4386004948 hasRelatedWork W2345479200 @default.
- W4386004948 hasRelatedWork W2419146053 @default.
- W4386004948 hasRelatedWork W2549990292 @default.
- W4386004948 hasRelatedWork W2849310602 @default.
- W4386004948 hasRelatedWork W2951819827 @default.
- W4386004948 hasRelatedWork W3006008237 @default.
- W4386004948 hasRelatedWork W4313153715 @default.
- W4386004948 hasRelatedWork W2183306018 @default.
- W4386004948 hasRelatedWork W3111219495 @default.
- W4386004948 isParatext "false" @default.
- W4386004948 isRetracted "false" @default.
- W4386004948 workType "book-chapter" @default.