Researchers at University College London, Sheffield and Pennsylvania have turned to an artificial intelligence to predict human rights court cases.
Using an AI method developed by researchers at the universities they were able to correctly predict 79% of the verdicts at the European Court of Human Rights.
Based on a dataset of 584 cases related to three articles of European Convention on Human Rights, an algorithm was used to look for patterns in text and then label each case as a violation or non-violation.
The cases were selected on three articles; Article 3, which involves cases related to torture or degrading treatment, Article 6, rights to a fair trial, and Article 8, respect for private life.
The researchers said that these were picked because there was both a large amount of published data on them and because they represent cases about fundamental rights.
The method is the first to use a machine learning algorithm to analyse case text, however, the researchers don’t expect to see AI replacing judges or lawyers.
Dr Nikolaos Aletras, who led the study at UCL Computer Science, said: “We don’t see AI replacing judges or lawyers, but we think they’d find it useful for rapidly identifying patterns in cases that lead to certain outcomes. It could also be a valuable tool for highlighting which cases are most likely to be violations of the European Convention on Human Rights.”
In order to train AI systems there is typically a lot of data required in order to get the most accurate answers. This is something that the researchers struggled with as they had to rely upon court-published summaries of applications made to the court, rather than the original applications themselves.
It was found that the most reliable factors for predicting the court’s decision were the language used and the topics and circumstances mentioned in the case text. The circumstances related to the factual background to the case.
The researchers combined information extracted from the abstract topics that the cases cover and circumstances across data for all three articles, this achieved an accuracy of 79%.
Dr Vasileios Lampos, UCL Computer Science, and co-author, said: “Previous studies have predicted outcomes based on the nature of the crime, or the policy position of each judge, so this is the first time judgements have been predicted using analysis of text prepared by the court.
“We expect this sort of tool would improve efficiencies of high level, in demand courts, but to become a reality, we need to test it against more articles and the case data submitted to the court.”
This kind of technology is increasingly being looked at across numerous areas as a way to augment human capabilities rather than to replace people, such as in financial services.
