This dataset was originally created in August 2018 and was published in Multi-class Multilingual Classification of Wikipedia Articles Using Extended Named Entity Tag Set in LREC'20 (https://huggingface.co/papers/1909.06502).

Description of the data files

enes.json: This file contains the annotation classes originally provided in the 'Extended Named Entity Hierarchy'. You can find more information about this in the 'Extended Named Entity Hierarchy' (http://www.lrec-conf.org/proceedings/lrec2002/pdf/120.pdf)
entity_annotations.json: This file includes hand annotations for Japanese Wikipedia pages along with Wikipedia page IDs of equivalent pages in English, French, German, and Farsi Wikipedia (wherever an equivalent page has been available).
data.jsonl: This standalone JSONL format file combines the information from enes.json and entity_annotations.json with downloaded Wikipedia articles. The articles were downloaded from Wikipedia during the last week of April 2024.
example_data_reader.py: This Python script provides an example data reader aimed at expediting the process of parsing and utilizing the data files.

If you have any questions or need assistance, feel free to reach out to Hassan S. Shavarani (sshavara@sfu.ca).