“CaseReview” An Integrated Approach for the Analysis And Suggestion of Sri Lankan Legal Case Documents

Abstract

Over recent times, there have been numerous technological innovations in the legal domain. The use of Artificial Intelligence to automate various tasks in the field of law is a growing trend. The application of Machine Learning and Natural Language Processing approaches to automate tasks in the legal domain is a well-established endeavor. Machine Learning and Natural Language Processing techniques have been extensively used for automating various tasks pertaining to the analysis and interpretation of legal texts. A legal case document is a peculiar type of a legal text. It contains the record of a particular judgment of a competent court made on a particular legal issue raised by one party or another. Legal case documents are used for reference by legal professionals, competent courts and other personnel in order to determine the precedent of legal issues. Initial literature survey has revealed that there is a dearth in work which explore the use of Natural Language Processing and Machine Learning in automating tasks pertaining to the analysis of legal case documents. This primary aim of this research project is to demonstrate the applicability of Natural Language Processing and Machine Learning techniques in the analysis of legal case documents. The most notable types of approaches which were explored were: Text classification, text summarization and information extraction. The proposed solution addresses the research gap of the lack of work exploring the use of Machine Learning and Natural Language Processing approaches in analyzing legal case documents or judgment records. Moreover, the proposed solution also introduces a novel integrated approach combining three techniques: Text Classification, Text Summarization and Information Retrieval.