NewsBench / README.md
m-ric's picture
Upload dataset
33f411d verified
metadata
dataset_info:
  features:
    - name: thread_uuid
      dtype: string
    - name: thread_url
      dtype: string
    - name: thread_site_full
      dtype: string
    - name: thread_site
      dtype: string
    - name: thread_site_section
      dtype: string
    - name: thread_site_categories
      list: string
    - name: thread_section_title
      dtype: string
    - name: thread_site_title
      dtype: string
    - name: thread_title
      dtype: string
    - name: thread_title_full
      dtype: string
    - name: thread_published
      dtype: string
    - name: thread_replies_count
      dtype: int64
    - name: thread_participants_count
      dtype: int64
    - name: thread_site_type
      dtype: string
    - name: thread_country
      dtype: string
    - name: thread_main_image
      dtype: string
    - name: thread_performance_score
      dtype: int64
    - name: thread_domain_rank
      dtype: float64
    - name: thread_domain_rank_updated
      dtype: string
    - name: thread_licensing_agency
      list: string
    - name: thread_social_facebook_likes
      dtype: int64
    - name: thread_social_facebook_comments
      dtype: int64
    - name: thread_social_facebook_shares
      dtype: int64
    - name: thread_social_vk_shares
      dtype: int64
    - name: uuid
      dtype: string
    - name: url
      dtype: string
    - name: ord_in_thread
      dtype: int64
    - name: author
      dtype: string
    - name: published
      dtype: string
    - name: title
      dtype: string
    - name: text
      dtype: string
    - name: highlightText
      dtype: string
    - name: highlightTitle
      dtype: string
    - name: highlightThreadTitle
      dtype: string
    - name: language
      dtype: string
    - name: sentiment
      dtype: string
    - name: categories
      list: string
    - name: topics
      list: string
    - name: ai_allow
      dtype: bool
    - name: has_canonical
      dtype: bool
    - name: breaking
      dtype: bool
    - name: webz_reporter
      dtype: bool
    - name: external_links
      list: string
    - name: entities
      dtype: string
    - name: syndication_syndicated
      dtype: bool
    - name: syndication_syndicate_id
      dtype: string
    - name: syndication_first_syndicated
      dtype: bool
    - name: trust_categories
      list: string
    - name: trust_top_news
      list: string
    - name: trust_bias
      dtype: string
    - name: rating
      dtype: 'null'
    - name: crawled
      dtype: string
    - name: updated
      dtype: string
    - name: file_path
      dtype: string
    - name: dataset_name
      dtype: string
    - name: sentiment_from_path
      dtype: string
    - name: source
      dtype: string
    - name: entities_json
      dtype: string
    - name: thread_json
      dtype: string
    - name: trust_source_type
      dtype: string
    - name: trust_source_city
      list: string
    - name: trust_source_state
      list: string
    - name: trust_source_country
      list: string
    - name: trust_source_domain_type
      dtype: string
    - name: trust_source_agency
      dtype: string
    - name: trust_source_organization_name
      dtype: string
    - name: parent_url
      dtype: float64
    - name: canonical
      dtype: bool
    - name: external_images
      list: 'null'
    - name: trust
      dtype: float64
    - name: internal_images
      list: 'null'
  splits:
    - name: train
      num_bytes: 686071011
      num_examples: 100000
  download_size: 301601913
  dataset_size: 686071011
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*