cicitar

  • Descrição :

Este é um conjunto de dados para classificar intenções de citação em trabalhos acadêmicos. O rótulo de intenção de citação principal para cada objeto Json é especificado com a chave de rótulo, enquanto o contexto de citação é especificado com uma chave de contexto. Exemplo:

{
 'string': 'In chacma baboons, male-infant relationships can be linked to both
    formation of friendships and paternity success [30,31].'
 'sectionName': 'Introduction',
 'label': 'background',
 'citingPaperId': '7a6b2d4b405439',
 'citedPaperId': '9d1abadc55b5e0',
 ...
 }

Você pode obter as informações completas sobre o papel usando os IDs de papel fornecidos com a API do Semantic Scholar ( https://api.semanticscholar.org/ ).

Os rótulos são: Método, Histórico, Resultado

Dividir Exemplos
'test' 1.859
'train' 8.194
'validation' 916
  • Estrutura de recursos :
FeaturesDict({
    'citeEnd': int64,
    'citeStart': int64,
    'citedPaperId': Text(shape=(), dtype=string),
    'citingPaperId': Text(shape=(), dtype=string),
    'excerpt_index': int32,
    'id': Text(shape=(), dtype=string),
    'isKeyCitation': bool,
    'label': ClassLabel(shape=(), dtype=int64, num_classes=3),
    'label2': ClassLabel(shape=(), dtype=int64, num_classes=4),
    'label2_confidence': float32,
    'label_confidence': float32,
    'sectionName': Text(shape=(), dtype=string),
    'source': ClassLabel(shape=(), dtype=int64, num_classes=7),
    'string': Text(shape=(), dtype=string),
})
  • Documentação do recurso:
Funcionalidade Aula Forma Tipo D Descrição
RecursosDict
citeFim tensor int64
citarInício tensor int64
citadoPapelId Texto fragmento
citandoPaperId Texto fragmento
trecho_índice tensor int32
Eu iria Texto fragmento
isKeyCitation tensor bool
rótulo ClassLabel int64
label2 ClassLabel int64
label2_confiança tensor float32
label_confiança tensor float32
Nome da Seção Texto fragmento
fonte ClassLabel int64
fragmento Texto fragmento
  • Citação :
@InProceedings{Cohan2019Structural,
  author={Arman Cohan and Waleed Ammar and Madeleine Van Zuylen and Field Cady},
  title={Structural Scaffolds for Citation Intent Classification in Scientific Publications},
  booktitle="NAACL",
  year="2019"
}