search

IndoNLP / nusa-crowd

A collaborative project to collect datasets in Indonesian languages.
Apache License 2.0
261 stars 61 forks source link

Create dataset loader for Multilingual Open Relations #305

Open SamuelCahyawijaya opened 1 year ago

SamuelCahyawijaya commented 1 year ago

NusaCatalogue: https://indonlp.github.io/nusa-catalogue/card.html?open_relation_id

Dataset open_relation_id
Description This dataset provides the set of automatically extracted relations obtained using cross-lingual annotation projection method. The data covers 61 languages, including Indonesian. Relation extraction is the task of assigning a semantic relationship between a pair of arguments. For example, from the sentence "Soekarno lahir di Jawa Timur", the relation <"Soekarno", born_in, "Jawa Timur"> is expected to be extracted.
License Unknown
madenindya commented 1 year ago

self-assign