Since the 1970s, the role of AI in law and legal service delivery has been a subject of research and development.
Propelled by more powerful computers and machines with greater capacity for storage, the last decade has seen an increased interest in the role of AI in law. Our research in AI focuses machine learning (including neural network/deep learning systems), quantitative legal prediction, and expert systems, applied to the practice of law. In conjunction with leading law firms we undertake industry-led research and development projects.
Automated Fact Checking
Project Lead: Dr Niall McCarroll
This project explores automated fact checking of legal documents using computational intelligence techniques where the aim is to extract and verify each fact in specific legal texts. Knowledge acquisition rules, based on the linguistic treatment of specific aspects of legal documents is key to improving the results in this task. Additionally, domain knowledge representation can provide an even broader set of possibilities.
This research will create language models for addressing Information Extraction from texts in the legal domain combined with external publicly accessible document silos in order to verify statements. Automatic fact checking of legal documents allows for improvements in legal information retrieval system effectiveness.
This project builds on Ulster University's prior research into automated subtitling and language identification where we developed hidden markov models, lexicons and phoneme bi/tri-gram sequences for any natural language modelled (e.g. English). A core outcome is language models generated from lexicons, grammars and phoneme databases with training on spoken dialogue data.