A group of scientist at University College London (UCL), the University of Sheffield and the University of Pennsylvania, has developed an Artificial Intelligence (A.I.) "judge" who would be able to predict the outcome of human right trials by analyzing case text using a machine learning algorithm.
Cases In The European Court Of Human Rights
According to the International Business Times, the study conducted by the researchers at these academic institutions showed that the A.I. system has predicted the outcome of hundreds of cases heard at the European Court of Human Rights with 79 percent accuracy. Although this is not precisely a bad number, the scientist explained that the machines wouldn't replace judges or lawyers.
"We don't see A.I. replacing judges or lawyers, but we think they'd be 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," lead researcher from UCL's department of computer science, Dr. Nikolaos Aletras, told to the media.
How The A.I. "Judges" Worked
According to the Daily Mail, the scientist looked at case information by the European Court of Human Rights (ECtHR) in its accessible database, identified English language data sets for more than 500 cases relating to the articles 3, 6, an 8 of the organization, and applied an A.I. algorithm to find pattern in the text.
Also, the A.I. "judges" had to analyze an equal number of violations and non-violations cases in order to prevent bias and mislearning. Apparently, the results showed that language, topics and the circumstances mentioned in the case text were the most reliable facto for predicting the court´s decision.
"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 judgments 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," UCL computer scientist Dr. Vasileios Lampos said.