Sigalei Analytics: Turning Regulatory Documents into Strategic Decisions

• Grant number: 23/16491-5
• Support Opportunities: Research Grants - Innovative Research in Small Business - PIPE
• Start date: July 01, 2024
• End date: June 30, 2026
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: Danilo Amaral de Oliveira
• Grantee: Danilo Amaral de Oliveira
• Company: Openlex Soluções Tecnológicas Ltda
• CNAE: Tratamento de dados, provedores de serviços de aplicação e serviços de hospedagem na internet ; Outras atividades de prestação de serviços de informação não especificadas anteriormente
• City: São Carlos
• Pesquisadores principais:( Atuais ): Frederico Amaral de Oliveira ; Ivan Ervolino
• Pesquisadores principais:( Anteriores ): Bruno Squizato Faiçal
• Associated research grant: 22/10596-7 - Regulatory document extraction engine (MEDoRe), AP.PIPE
• Associated scholarship(s): 24/09655-4 - Sigalei analytics: turning regulatory documents into strategic decisions, BP.PIPE; 24/09221-4 - Data modeling formalization and infrastructure adaptation for processed and non processed documents storage and availability, BP.TT; 24/09245-0 - Collection of regulatory documents, BP.TT; + associated scholarships 24/09182-9 - Generation of customizable analyses from actionable data (MEDoRe - Analysis), BP.TT; 24/09212-5 - Transforming unstructured documents into actionable data (MEDoRe - Extraction), BP.TT; 24/09211-9 - Collection of complementary data for customizable analyses., BP.TT - associated scholarships

Abstract

Government decisions have a significant impact on the economic sector, bringing risks and opportunities. Companies therefore need to develop a process for monitoring and analyzing government information, taking into account the reality of their areas of activity, their value propositions and their operations in order to effectively address the risks and opportunities for their business. One of the main barriers is the vast and variable volume of regulatory data generated on a daily basis, which is often complex and difficult to understand. Aggravating the scenario is the fact that much of the data made available is unstructured, i.e. it cannot be analyzed using computational methods.In Phase 1 of PIPE, the Regulatory Document Extraction Engine (MEDoRe) was created with the aim of structuring acts from municipal and state official gazettes in a scalable way. It was discovered that MEDoRe can structure not only official gazettes, but also other regulatory documents such as bills and contracts, broadening its applicability. In addition, from an application case developed, it was possible to validate the commercial interest and the ability to generate personalized information and analysis. It is important to note that the use of MEDoRe made it possible to carry out a task that was considered unfeasible for specialists to do manually due to the large volume of documents, making it automated and updatable on demand.This project aims to integrate MEDoRe into Sigalei, giving rise to Sigalei Analytics, based on the methodology and team proven in Phase 1. MEDoRe will be divided into two modules MEDoRe - Extraction and MEDoRe - Analysis. The first will be carried out continuously and automatically for all the regulatory documents collected. The second consists of a base of pre-trained machine learning models that will be parameterized and customized according to the specific demands of each client. Both modules interact with each other and with complementary modules to create Sigalei Analytics, a platform that will enable analysis and the generation of information based on actionable data from the structuring of different regulatory documents to effectively meet the identified demand. (AU)