trec

  • Descrizione :

Il set di dati di classificazione delle domande della Text REtrieval Conference (TREC) contiene 5500 domande etichettate nel set di addestramento e altre 500 per il set di test. Il set di dati ha 6 etichette, 47 etichette di livello 2. La lunghezza media di ogni frase è 10, la dimensione del vocabolario di 8700. I dati sono raccolti da quattro fonti: 4.500 domande in inglese pubblicate da USC (Hovy et al., 2001), circa 500 domande costruite manualmente per alcune classi rare, 894 TREC 8 e TREC 9 domande e anche 500 domande da TREC 10 che funge da set di test.

Diviso Esempi
'test' 500
'train' 5.452
  • Struttura delle caratteristiche :
FeaturesDict({
    'label-coarse': ClassLabel(shape=(), dtype=int64, num_classes=6),
    'label-fine': ClassLabel(shape=(), dtype=int64, num_classes=47),
    'text': Text(shape=(), dtype=string),
})
  • Documentazione delle funzionalità :
Caratteristica Classe Forma Tipo D Descrizione
CaratteristicheDict
etichetta grossolana ClassLabel int64
etichetta-bene ClassLabel int64
testo Testo corda
  • Citazione :
@inproceedings{li-roth-2002-learning,
    title = "Learning Question Classifiers",
    author = "Li, Xin  and
      Roth, Dan",
    booktitle = "{COLING} 2002: The 19th International Conference on Computational Linguistics",
    year = "2002",
    url = "https://www.aclweb.org/anthology/C02-1150",
}
@inproceedings{hovy-etal-2001-toward,
    title = "Toward Semantics-Based Answer Pinpointing",
    author = "Hovy, Eduard  and
      Gerber, Laurie  and
      Hermjakob, Ulf  and
      Lin, Chin-Yew  and
      Ravichandran, Deepak",
    booktitle = "Proceedings of the First International Conference on Human Language Technology Research",
    year = "2001",
    url = "https://www.aclweb.org/anthology/H01-1069",
}