README.md

metadata

task_categories:
  - text-classification
language:
  - en
tags:
  - legal
size_categories:
  - 10K<n<100K

This is the ECHR dataset, a collection of 11.5K court cases extracted from the public database of the European Court of Human Rights and further annotated by human experts. The dataset was published along with this paper (pleae cite it accordingly!) and can be donwloaded in its original form from this website.

Each instance in this dataset is a court case. Each court case is annotated with the following properties (the columns of the dataframe):

• partition: a label indicating dataset partition this court case belongs to ("train", "dev", or "test")
• itemid: a code which uniquely identifies this court case
• languageisocode: an ISO code describing the language in which the case is reported
• respondent: the ISO code of the party being sued or tried (respondents are nation states)
• branch: the branch of the Court dealing with the case, indicating at which stage of the trial a judgement was made (it can be one out of "ADMISSIBILITY", "CHAMBER", "GRANDCHAMBER", "COMMITTEE")
• date: the date of the judgement
• docname: the title of the court case (for example, "ERIKSON v. ITALY")
• importance: an "importance score" from 1 (key case) to 4 (unimportant), denoting a case's contribution in the development of case-law
• conclusion: a short summary of the case conclusion (for example, "Inadmissible" or "Violation of Art. 6-1; No violation of Art. 10"
• judges: the name of the judges
• text: the facts brought to the attention of the Court
• binary_judgement: a binary label indicating whether an article or protocol was (1) or wasn't (0) violated