README.md

metadata

language:
  - en
license: apache-2.0
dataset_info:
  features:
    - name: text
      dtype: string
    - name: label
      dtype: int64
    - name: bert_embeddings
      sequence: float32
    - name: roberta_embeddings
      sequence: float32
    - name: user_interactions
      list:
        - name: content
          dtype: string
        - name: id
          dtype: string
        - name: tone
          dtype: string
    - name: interaction_embeddings_list
      sequence:
        sequence: float64
    - name: interaction_tones
      sequence: string
    - name: combined_embeddings
      sequence: float64
    - name: distilbert_embeddings
      sequence: float32
    - name: bigbird_embeddings
      sequence: float32
    - name: deberta_embeddings
      sequence: float32
  splits:
    - name: train
      num_bytes: 1584446149
      num_examples: 9988
    - name: test
      num_bytes: 422704060
      num_examples: 2672
  download_size: 1608156828
  dataset_size: 2007150209
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*

Dataset Source: Ahren09/MMSoc_GossipCop

This is a copied and reformatted version of the Ahren09/MMSoc_GossipCop

• text: text of the article (str)
• bert_embeddings: (768, )
• roberta_embeddings: (768, )
• label: (int)
  • 0: real
  • 1: fake

Datasets Distribution:

• Train: 9988 (real: 7955, fake: 2033)
• Test: 2672 (real: 2169, 503)