Datasets:

boragokbakan
/

entity_disambiguation

Tasks:

Languages:

Tags:

License:

Dataset Viewer

The dataset viewer is not available for this dataset.

Unexpected token '<', "<html> <h"... is not valid JSON

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

Entity Disambiguation datasets as provided in the GENRE repo. The dataset can be used to train and evaluate entity disambiguators.

The datasets can be imported easily as follows:

from datasets import load_dataset

ds = load_dataset("boragokbakan/entity_disambiguation", "aida")

Available dataset names are:

Note: As the BLINK training set is very large in size (~10GB), it is advised to set streaming=True when calling load_dataset.