Case brief generation and verdict prediction using natural language processing and machine learning

Abstract

The paper's objective was to demonstrate the potential of natural language processing and machine learning in solving the challenges faced by the Kenyan judicial system, particularly the increasing complexity and the large backlog of cases. To achieve this, the research utilised NLP and AI algorithms to automatically generate case briefs with relevant precedent cases and predict the likely outcome of the verdict. Selected lawyers provided expert labelling of downloaded cases and sample legal case briefs. The labelled data was then used to train the model. The resulting toolkit was capable of generating case briefs and predicting verdicts with impressive levels of accuracy, 88% for case briefs and 83.7% for verdict predictions. The findings of the research demonstrate that NLP and ML is a valuable solution that can enhance human abilities by automating case brief generation and verdict prediction and potentially. This will contribute to more efficient service delivery.