Automatic knowledgebase representation of legal data

Bailung, Naksatra Kumar (2022) Automatic knowledgebase representation of legal data. Masters thesis, Indian Institute of Science Education and Research Kolkata.

Text (MS dissertation of Naksatra Kumar Bailung (17MS032)) 17MS032_Thesis_file.pdf - Submitted Version Restricted to Repository staff only Download (4MB)

Abstract

The large influx of data in the new generation has invited many professionals to start researching the field of data representation. While many still consider it a nascent industry, people are still hopeful and amazed by the use of computer algorithms on these huge datasets to infer results that happen in the real world. One should remember that the data obtained is usually extremely unstructured or riddled with errors, hence cleaning the data becomes an integral part of accessing the data. Even after cleaning this data, we obtain correctly syntactic files that again are pure information and weird to traverse. Consumers prefer to solve issues on their own; hence it becomes the duty of the knowledge engineers to provide them with an interface that can answer their queries with high precision. One aims to create a knowledge representation that can cater to the user’s needs about a specific domain. Hence knowledge-based systems are designed to emulate the work of an expert in a specific domain of knowledge. These systems comprise the complete knowledge of the domain in a compressed format and allow easy access for the users to use and revel in. In this project, we focus on concepts generated from Supreme court cases, prioritising the ones occurring more in cases with Indian Penal Code (IPC) sections. We generate ontologies according to and study their properties and impact on the data using these concepts. Primarily focusing on text that appear more apparently in court cases involving the mention of IPC cases, we hope that this generated database can be used to resolve user queries and cases extremely fast and provide an alternate mechanism for lawyers to take insights into previous court cases.

Actions (login required)

View Item