Context and overall objectives

Project ADELE is premised on the ongoing paradigm shift towards cognitive computing and human-centred AI which is transforming many socio-economic activities, including justice. The project applies legal analytics (LA) – a blend of data science, machine learning and natural language processing techniques – to judicial decisions: it aims to(1) develop methods to extract knowledge and engage in outcome predictions and therefrom (2) build a pilot tool to support legal research and decision-making processes in the judiciary.

Work planned

The project will begin with the analysis of four datasets of judgments respectively of Italian and Bulgarian case law in the fields of Intellectual Property and Value Added Tax (WP2). Annotation guidelines will be developed and used to tag the datasets. Domain-specific ontologies and citation networks will be built.

In WP3 tagged documents will be processed with machine learning algorithms to extract legal arguments and predict the possible outcome of new cases. Two German datasets will be collected and used as a language transferability test.

WP4 will develop a pilot tool embedding knowledge extraction and outcome prediction functionalities.

Results and Impact

The tool will be subject to (1) a preliminary test on a beta version of the tool; (2) a validation test on the final pilot (WP5). Tests will take place both in Bulgaria and Italy with specialized end-users. The main outputs of ADELE will be (1) a LA methodological framework and (2) a pilot tool to support the activities of justices in the Italian and Bulgarian courts. These novel solutions will have a strong impact on the efficacy and transparency of judicial decision-making by developing automated means that enhance the understanding of the law, facilitate judicial tasks and reduce the courts’ workload. It will contribute to developing a trustworthy AI by producing ethical and robust LA methods for the judiciary and will foster scientific debate on the use of AI technologies in the legal field.

Here below are more details about project articulation per work package. Go to the Agenda to check out the project timeline and milestones.