story_cloze

  • Açıklama :

Story Cloze Test, hikaye anlayışını, hikaye oluşturmayı ve senaryo öğrenimini değerlendirmek için yeni bir sağduyulu muhakeme çerçevesidir. Bu test, dört cümlelik bir hikayenin doğru sonunu seçmek için bir sistem gerektirir.

  • Ek Belgeler : Belgeleri Keşfedin

  • Yapılandırma açıklaması : 2018 yılı

  • Ana sayfa : https://www.cs.rochester.edu/nlp/rocstories/

  • Kaynak kodu : tfds.datasets.story_cloze.Builder

  • sürümler :

    • 1.0.0 (varsayılan): İlk sürüm.
  • İndirme boyutu : Unknown size

  • Manuel indirme talimatları : Bu veri kümesi, kaynak verileri manuel olarak download_config.manual_dir içine download_config.manual_dir gerektirir (varsayılan olarak ~/tensorflow_datasets/downloads/manual/ ):
    Veri kümelerini edinmek için https://www.cs.rochester.edu/nlp/rocstories/ adresini ziyaret edin ve google formunu doldurun. Veri kümelerini indirmek için bağlantı içeren bir e-posta alacaksınız. 2016 verileri için doğrulama ve test dosyasının sırasıyla cloze_test val _spring2016.csv ve cloze_test test _spring2016.csv olarak yeniden adlandırılması gerekir. 2018 sürümü için, doğrulama ve test dosyasının sırasıyla cloze_test val _winter2018.csv ve cloze_test test _winter2018.csv olarak yeniden adlandırılması gerekir. Bu dosyaların ikisini de kılavuz dizinine taşıyın.

  • Otomatik önbelleğe alınmış ( belgeleme ): Evet

  • Özellik yapısı :

FeaturesDict({
    'context': Text(shape=(), dtype=string),
    'endings': Sequence(Text(shape=(), dtype=string)),
    'label': int32,
})
  • Özellik belgeleri :
Özellik Sınıf Şekil Dtipi Açıklama
ÖzelliklerDict
bağlam Metin sicim
sonlar Sıra(Metin) (Hiçbiri,) sicim
etiket tensör int32
@inproceedings{sharma-etal-2018-tackling,
    title = "Tackling the Story Ending Biases in The Story Cloze Test",
    author = "Sharma, Rishi  and
      Allen, James  and
      Bakhshandeh, Omid  and
      Mostafazadeh, Nasrin",
    booktitle = "Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)",
    month = jul,
    year = "2018",
    address = "Melbourne, Australia",
    publisher = "Association for Computational Linguistics",
    url = "https://www.aclweb.org/anthology/P18-2119",
    doi = "10.18653/v1/P18-2119",
    pages = "752--757",
    abstract = "The Story Cloze Test (SCT) is a recent framework for evaluating story comprehension and script learning. There have been a variety of models tackling the SCT so far. Although the original goal behind the SCT was to require systems to perform deep language understanding and commonsense reasoning for successful narrative understanding, some recent models could perform significantly better than the initial baselines by leveraging human-authorship biases discovered in the SCT dataset. In order to shed some light on this issue, we have performed various data analysis and analyzed a variety of top performing models presented for this task. Given the statistics we have aggregated, we have designed a new crowdsourcing scheme that creates a new SCT dataset, which overcomes some of the biases. We benchmark a few models on the new dataset and show that the top-performing model on the original SCT dataset fails to keep up its performance. Our findings further signify the importance of benchmarking NLP systems on various evolving test sets.",
}

story_cloze/2016 (varsayılan yapılandırma)

  • Veri kümesi boyutu : 1.15 MiB

  • bölmeler :

Bölmek örnekler
'test' 1.871
'validation' 1.871

story_cloze/2018

  • Veri kümesi boyutu : 1015.04 KiB

  • bölmeler :

Bölmek örnekler
'test' 1.571
'validation' 1.571