Corpus for Automatic Structuring of Legal Documents

Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022), pages 4420–4429 Marseille, 20-25 June 2022 © European Language Resources Association (ELRA), licensed under CC-BY-NC-4.0 4420 Corpus for Automatic Structuring of Legal Documents Prathamesh Kalamkar1,2,∗, Aman Tiwari1,2,∗, Astha Agarwal1,2,∗, Saurabh Karn3,∗, Smita Gupta3, Vivek Raghavan1, Ashutosh Modi4 1EkStep Foundation, 2Thoughtworks Technologies India Pvt Ltd., 3Agami , 4 Indian Institute of Technology Kanpur (IIT-K) {prathamk, aman.tiwari, astha.agarwal}@thoughtworks.com, {saurabh, smita}@agami.in, vivek@ekstep.org, ashutoshm@cse.iitk.ac.in Abstract In populous countries, pending legal cases have been growing exponentially. There is a need for developing techniques for processing and organizing legal documents. In this paper, we introduce a new corpus for structuring legal documents. In particular, we introduce a corpus of legal judgment documents in English that are segmented into topical and coherent parts. Each of these parts is annotated with a label coming from a list of pre-defined Rhetorical Roles. We develop baseline models for automatically predicting rhetorical roles in a legal document based on the annotated corpus. Further, we show the application of rhetorical roles to improve performance on the tasks of summarization and legal judgment prediction. We release the corpus and baseline model code along with the paper. Keywords: Legal NLP, Rhetorical Roles, Legal Document Segmentation 1. Introduction In populous countries (e.g., India), pending legal cases have been growing exponentially. For example, ac- cording to India’s National Judicial Data Grid, as of December 2021, there are approximately 40 million cases pending in various courts of the country (National Judicial Data Grid, 2021). India follows a common-law system; consequently, due to subjectivity involved in the legal process, it may not be possible to automate the entire judicial pipeline completely; nevertheless, many intermediate tasks can be automated to augment legal practitioners, and hence expedite the system. For ex- ample, legal documents can be processed with the help of Natural Language Processing (NLP) techniques to organize and structure the data to be amenable to au- tomatic search and retrieval. However, legal texts are different from commonly occurring texts typically used to train NLP models. Legal documents are quite long, running into tens (sometimes hundreds) of pages. Long documents make automatic processing challenging as information is spread throughout the document (Ma- lik et al., 2021b). Another challenge with legal doc- uments is the use of different lexicons. Though legal documents use natural language (e.g., English), many commonly occurring words/terms have different legal connotations. The use of different lexicons makes it challenging to adapt existing NLP models to legal texts (Malik et al., 2021b). Moreover, in countries like In- dia, legal documents are manually typed and are highly unstructured and noisy (e.g., spelling and grammatical mistakes). Above mentioned challenges make it dif- ficult to apply existing NLP models and techniques di- rectly, which calls for the development of legal domain- specific techniques. Existing state-of-the-art models in NLP are data-driven * Authors contributed equally and are trained on annotated corpora. However, the legal domain suffers from the deficiency of availabil- ity of annotated corpora. It has hindered the growth …

Content truncated to 500 words