---
dataset_info:
  features:
  - name: _id
    dtype: string
  - name: character_id
    dtype: string
  - name: article_en
    dtype: string
  - name: article_vi
    dtype: string
  - name: character_url
    dtype: string
  - name: code
    dtype: string
  - name: description
    dtype: string
  - name: language
    dtype: string
  - name: title
    dtype: string
  - name: type
    dtype: string
  - name: metadata
    dtype: string
  - name: summary
    dtype: string
  - name: text
    dtype: string
  splits:
  - name: train
    num_bytes: 5843304840
    num_examples: 515669
  download_size: 3190490993
  dataset_size: 5843304840
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---
This dataset is used by me using wiki data to get all characters in English and Vietnamese pages in the following list of categories:
- Q5 : human
- Q95074 : fictional_character
- Q15632617 : fictional_human
- Q15773347 : television_character
- Q1114461 : comic_book_character
- Q4271324 : literary_character

Then I use wikipedia-api to get the content of each page characters