Alice Balanço Q&A: question and answer system for financial services institution

• Grant number: 23/09956-1
• Support Opportunities: Research Grants - Innovative Research in Small Business - PIPE
• Duration: June 01, 2024 - May 31, 2026
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Convênio/Acordo: SEBRAE-SP
• Principal Investigator: Valdirene Fontanette
• Grantee: Valdirene Fontanette
• Host Company: Itera Inovação e Desenvolvimento Tecnológico Ltda. - ME
• CNAE: Desenvolvimento e licenciamento de programas de computador não-customizáveis ; Suporte técnico, manutenção e outros serviços em tecnologia da informação ; Tratamento de dados, provedores de serviços de aplicação e serviços de hospedagem na internet
• City: São Carlos
• Pesquisadores principais: Renato Lewenthal Carrião
• Associated researchers: Solange Oliveira Rezende
• Associated research grant: 20/05895-0 - A question answering system for financial economic domain, AP.PIPE
• Associated grant(s): 24/04244-6 - Alice Balanço Q&A: question and answer system for financial services institution, AP.PIPE

Abstract

Itera aims to develop technologies related to Artificial Intelligence (AI), especially those related to the use of Text Mining, Natural Language Processing (NLP) and Machine Learning (ML) techniques, to solve real market problems. The services provided by Itera have, as their main differential, the speed in terms of time of development, instantiation and operationalization of AI solutions. In this sense, the main product of Itera's portfolio today is Alice Balanço, a service that, through the automatic extraction of accounting balance sheets, allows a personalized configuration for each client, helping to automate part of the back office of manual operations of credit risk analysis made by financial service institutions that extend credit to companies. For financial institutions to calculate the score of companies that demand their credit offers, they analyze, among various documents, the social contract of these companies. Thus, it was identified that customers face many challenges in recovering information from this type of document, as they do not have mechanisms to automate the recovery of this information, being a purely manual and costly activity. In this context, a Question and Answer System can meet this need for extracting information quickly and objectively, automatically providing answers to questions from credit risk analysts about social contracts. Thus, the objective of this project is to promote the innovation of Alice Balanço with a Question and Answer system for the financial area, Alice Balanço Q&A, offering: an environment for extraction and management of information from balance sheets and social contracts of companies that apply for credit, enabling the analysis of the company to be carried out faster and more efficiently. This environment, as a product, will allow new ways of exploring information, enabling a better understanding and improving the quality of the credit that is given.In view of the proof of the feasibility of using the Extractive Approach for a Question and Answer System researched in the PIPE-Phase I project, it is now kept in mind that new investments must be made to improve the prototype that was built, researching and comparing the Approach Extractive with other possible approaches to solve the question and answer domain problem, investigating new and better techniques for visualization of the generated data, enabling the prototype to process large volumes of data in a Big Data environment with scalable software architecture in the cloud. (AU)