README.md

metadata

dataset_info:
  features:
    - name: en
      dtype: string
    - name: ar
      dtype: string
  splits:
    - name: train
      num_bytes: 82235661
      num_examples: 818354
  download_size: 22311870
  dataset_size: 82235661
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

from datasets import load_dataset,Dataset

ds = load_dataset("Ibrahemqasim/categories_en2ar")

enwiki_to_arwiki_categories Dataset

This dataset contains mappings between English Wikipedia categories and their corresponding Arabic Wikipedia categories.

Files

1. langlinks.json (818,354)

   • This file contains all category links from enwiki to arwiki.
   • Dataset at: Ibrahemqasim/categories_en2ar

2. filtered_data.json (231,314)

   • This file contains the mappings after filtering out those that do not contain a 4-digit year.
   • Dataset at: Ibrahemqasim/categories_en2ar_with_years

3. cats_2000.json (21,170)

   • This file contains the mappings from filtered_data.json with these changes:
     1. Replacing all 4-digit years with the year 2000.
   • Dataset at: Ibrahemqasim/categories_en2ar-cats_2000

4. cats_2000_country.json (1,234)

   • This file contains the mappings from filtered_data.json with these changes:
     1. Replacing all 4-digit years with the year 2000.
     2. Replacing country names with the word country. It contains 1,234 mappings.
   • Dataset at: Ibrahemqasim/categories_en2ar-cats_2000_contry