Exploring Resourcefulness of the Paragraph for PrecedenceRetrieval in Legal Documents

Abstract Developing computational methods for efficiently extracting relevant legal information to aid legalpractitioners is an active research area. In this regard, research efforts are being made to developefficient methods by leveraging different kinds of information in a legal document, such as meta-data, citations, keywords, sentences, paragraphs, etc. Similar to any text document, legal documentsare composed of paragraphs. In this paper, we have analyzed the resourcefulness of paragraph-levelinformation in capturing similarity among judgments for improving the performance of precedenceretrieval. The statistical analysis on the Supreme Court of India judgment dataset shows that theparagraph-level methods with only a few paragraph interactions could capture the similarity amongthe judgments and exhibit more discriminating power over the baseline document-level method. More-over, the comparison results on two benchmark datasets for the precedence retrieval on the Indiansupreme court judgments task show that the paragraph-level methods exhibit comparable perfor-mance with the state-of-the-art methods by utilizing only a few paragraph interactions. Overall, theresults show that paragraph-level granularity can better represent the legal document, and there is ascope to exploit them to improve the retrieval systems in the legal domain.