Legal Document Clustering and Summarization

Abstract

The massive increase in reliance on the Internet is paralleled by an increase in document digitization. With this expansion has come the need for rapid knowledge management, which is an automated tool for organizing, analyzing, retrieving, and displaying content in a meaningful and distributed manner. The justice system has been prevalent around the world for a long time, increasing the number of resources available for and in this field is a perpetual process. One such addition is the introduction of a legal search engine of sorts that allows for easy access to relevant documents for the user. Similarly, the precedent retrieval process has been merely based on references to past cases. However, there is a need to retrieve precedents irrespective of the references and on the basis of similar context. Hence, a context-aware similar document retrieval system has been proposed. Additionally, legal documents could be inscrutable if being accessed by someone who is not a part of the legal field. Therefore, there is a need for automatic summarizing documents in a way that can be understood by the common man. To tackle this problem, a summarization methodology has been implemented and its performance has been evaluated. Considering all these factors, a system needs to be developed with the following features: A search engine for legal documents which will provide the user with all the relevant documents they require along with their summaries. The system will also display recommendations of well-known lawyers in the subject of the issue the user is researching.