Advanced search
Start date
Betweenand

Law2Vec: a legal document writing aid tool using artificial intelligence

Grant number: 18/01099-4
Support type:Research Grants - Innovative Research in Small Business - PIPE
Duration: May 01, 2020 - January 31, 2021
Field of knowledge:Physical Sciences and Mathematics - Computer Science
Principal Investigator:Fernando José Vieira da Silva
Grantee:Fernando José Vieira da Silva
Company:F J V DA SILVA TECNOLOGIA INFORMACAO LTDA
CNAE: Desenvolvimento e licenciamento de programas de computador não-customizáveis
Tratamento de dados, provedores de serviços de aplicação e serviços de hospedagem na internet
City: Sorocaba

Abstract

The information technology adoption has being growing in law offices and companies' legal departments all around the Brazil. Currently, it is common to find systems for following litigation status, manage agenda and customer records, among other daily basis organizational process.However, there is still no specific available tool for helping the writing of petitions, contracts and legal documents in general, beyond traditional text editors such as Microsoft Word and alike. Nevertheless, our preliminary interviews with lawyers point that 85% of them spend at least half of their time writing such documents. Therefore, with the advent of technologies such as artificial intelligence, more specifically: machine learning and natural language processing, it opens a new horizon for creating new applications that aims at increasing productivity and quality on the production of legal documents. Therefore, we propose a new tool named Law2Vec that will be integrated into a text editor (as a plugin) and will use modern natural language processing techniques with the objective of increasing quality and efficiency on legal documents writing.In the project's first phase, we intend to approach the problem of legal case automatic search and citation, pointed by lawyers as of the most laborious tasks in the process. Although there are research in the legal case search field, the practical application remains little studied, once it also implies on scaling computational resources as for making such a system viable. Furthermore, this research also propose to innovate on creating a wide and updatable base of documents from legal decisions from several Brazilian courts. (AU)