In common law jurisdictions, legal professionals cite facts and legal prin- ciples from precedent cases to support their arguments before the court for their in- tended outcome in a current case. This practice stems from the doctrine of stare decisis, where cases that have similar facts should receive similar decisions with respect to the principles. It is essential for legal professionals to identify such facts and principles in precedent cases, though this is a highly time intensive task. In this paper, we present studies that demonstrate that human annotators can achieve reasonable agreement on which sentences in legal judgements contain cited facts and principles (respectively, κ = 0.65 and κ = 0.95 for inter- and intra-annotator agreement). We further demonstrate that it is feasible to automatically annotate sentences containing such legal facts and principles in a supervised machine learning framework based on linguistic features, reporting per category precision and recall figures of between 0.79 and 0.89 for classifying sentences in legal judgements as cited facts, principles or neither using a Bayesian classifier, with an overall κ of 0.72 with the human-annotated gold standard.
Bibliographical noteOpen access via Springer Compact Agreement
- natural language processing
- legal text analysis
- legal judgements