Researchers at the University of Bern in Switzerland have created a dataset to improve AI systems in the legal industry. The dataset contains over 130,000 sentences from laws and case judgements written in French, Italian, Spanish, Portuguese, English, and German as well as annotations describing each sentence’s structure. Researchers can use the dataset to train AI systems to process sentences in legal texts, which typically contain complex or abnormal structures.
Image credit: Flickr user Shrinivas Sankaran