скицит

  • Описание :

Это набор данных для классификации целей цитирования в научных статьях. Метка основного намерения цитирования для каждого объекта Json указывается с помощью ключа label, а контекст цитирования указывается с помощью ключа контекста. Пример:

{
 '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',
 ...
 }

Вы можете получить полную информацию о документе, используя предоставленные идентификаторы документов с API-интерфейсом Semantic Scholar ( https://api.semanticscholar.org/ ).

Ярлыки: Метод, Фон, Результат.

Расколоть Примеры
'test' 1859
'train' 8194
'validation' 916
  • Структура функции :
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),
})
  • Документация по функциям :
Особенность Учебный класс Форма Dтип Описание
ОсобенностиDict
citeEnd Тензор int64
citeStart Тензор int64
процитированныйPaperId Текст нить
цитированиеPaperId Текст нить
excerpt_index Тензор int32
я бы Текст нить
isKeyCitation Тензор логический
этикетка Метка класса int64
метка2 Метка класса int64
label2_confidence Тензор поплавок32
label_confidence Тензор поплавок32
sectionName Текст нить
источник Метка класса int64
нить Текст нить
  • Цитата :
@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"
}