The legal domain represents a primary candidate for Web-based information distribution, exchange, and management, and is motivated by numerous e-government, e-justice, and e-democracy initiatives worldwide. In the local sphere, Government provides publicly available legal information (in Maltese) such as judgements together with keyword-based search facilities available through the Justice ¸£ÀûÔÚÏßÃâ·Ñ, and more recently the eCourts portal.
Nevertheless, according to the EU Justice Scoreboard report of 2022, the Maltese court system is lagging other EU counterparts regarding the use of AI and the availability of machine-readable judgements to improve efficiency.
Recent years have seen an increase in research and practice in the field of Artificial Intelligence and Law intended to address aspects such as automated legal reasoning and argumentation, semantic and cross-language legal information retrieval, document classification, legal drafting and legal knowledge discovery and extraction, as well as predicting the outcome of legal cases.
AMPS is a joint initiative between the Faculty of Laws (Department of Public Law), and the Faculty of ¸£ÀûÔÚÏßÃâ·Ñ & Communication Technology (Department of Artificial Intelligence).
The scope behind this project is that of investigating the use of Natural Language Processing and Machine Learning to predict the outcome of Maltese court cases, specifically those within the Small Claims Tribunal, which deals with small claims, of up to €5000.
Although the nature of the cases in this tribunal differs from those of higher courts, the structure of the judgements is similar, albeit being less complex, thus allowing the team to investigate aspects related to case categorisation, legal arguments, and reasoning behind the judgement.
Since Maltese is still considered to be a low-resource language the project will consider recent research advances related to the development of Maltese corpora to build a newly specialised legal corpus comprising all the final rulings from the Small Claims Tribunal with the support of a team of Legal Experts.
AMPS will leverage on existing research in the areas of Natural Language Processing and Machine Learning including recent advances that focused specifically on building models that have been pre-trained on a multilingual and monolingual model for Maltese. The project will furthermore consider the issue of bias. Critics of legal supporting tools being introduced in courtrooms as part of several digital transformation initiatives, point to the limitations and the reasoning bias in available software. In this regard, this project will strive to be in line with the Council of Europe’s Ethical Charter for the use of AI in Judicial Systems and their Environmentand will adopt the evaluation checklist made available in this report.
Finally, the team intends to design, implement, and evaluate a proof of concept of the AMPS legal case prediction application. In this regards a focus group of legal experts and practitioners will be used to get feedback and improve the application such that it supports the Small Claims Tribunal adjudicators in their work.
AMPS Team at UM:
- Dr Ivan Mifsud (Faculty of Law)
- Dr Charlie Abela (Faculty of ICT)
- Dr Joel Azzopardi (Faculty of ICT)
- Mr Dawson Camilleri (Faculty of ICT)
