Knowledge base generation for customer service automation

• Grant number: 19/08609-0
• Support Opportunities: Research Grants - Innovative Research in Small Business - PIPE
• Start date: June 01, 2020
• End date: February 28, 2021
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Victor Hochgreb de Freitas
• Grantee: Victor Hochgreb de Freitas
• Company: Gobots Soluções Inteligentes Ltda
• CNAE: Desenvolvimento e licenciamento de programas de computador customizáveis ; Desenvolvimento e licenciamento de programas de computador não-customizáveis
• City: Santo André
• Pesquisadores principais: Julio Cesar dos Reis
• Associated scholarship(s): 20/09065-1 - Infrastructure for persisting knowledge graphs, BP.TT; 20/09711-0 - Questions and Answers processing about knowledge graphs, BP.TT; 20/06906-5 - Entities extraction to support the generation of knowledge graphs, BP.TT

Abstract

Effective and automated customer service done by computer systems requires the processing of large amounts of unstructured information. In online commerce customers ask questions about products and orders using natural language. The generation of automated answers to these requests demands the study of advanced techniques of natural language interpretation. GoBots develops high technology products to automatically answer questions about the most diverse products on online shopping platforms. This project aims to study, develop and validate solutions for automatic responses generation in intelligent customer service systems. The project aims to investigate and develop innovative techniques of natural language text semantic processing for the construction of knowledge graphs from text messages processing. Expected results involve the validation of a new software module capable of generating responses considering the knowledge present in existing messages history. (AU)