Sentence Extraction for Legal Text Summarisation

Ben Hachey, Claire Grover

Abstract: We describe a system for generating extractive summaries of texts in the legal domain, focusing on the relevance classifier, which determines which sentences are abstract-worthy. We experiment with naive Bayes and maximum entropy estimation toolkits and explore methods for selecting abstract-worthy sentences in rank order. Evaluation using standard accuracy measures and using correlation confirm the utility of our approach, but suggest different optimal configurations.