Book Title: Computational Criminology: AI Applications in Forensic Science and Criminal Justice
Editors: Dr. Xavier Louis, Dr. Surbhi Girdhar, Ms. Aswathi Chandran Nair, Mr. Ravi Kumar, and Ms. Nandini Katare
Chapter: 25
DOI: https://doi.org/10.59646/704/25
Author: Anjali M S
Abstract
The processing of legal text in its volume, terminological density, jurisdictional specificity, and interpretive complexity represents one of the most demanding and consequential natural language processing challenges. The digitisation of legal records, court opinions, statutes, and regulatory materials at scale has created a vast computational resource that NLP methods are increasingly able to exploit for practical legal applications: automated document review in discovery, intelligent case summarisation for judicial staff and appellate courts, precedent retrieval and citation recommendation, contract analysis and risk identification, and the emerging field of judicial analytics. This chapter surveys the NLP architectures and pre-training strategies that have proven most effective for legal text including Legal-BERT, LegalBench, and domain-adapted generative models evaluates their performance on key legal tasks, examines the human-in-the-loop workflow considerations essential to their responsible deployment, and addresses the access to justice implications of NLP tools that disproportionately benefit well-resourced litigants.