Author(s)

Thiago Henrique Barbosa de Carvalho Tavares¹, Bruno Pérez Ferreira², Eduardo Mazoni Andrade Marçal Mendes³

¹Departamento de Engenharia de Controle e Automação, Instituto Federal de Minas Gerais, Ibirité, Brasil.
²Departamento de Ciências Administrativas, Faculdade de Ciências Econômicas, Universidade Federal de Minas Gerais, Belo Horizonte, Brasil.
³Departamento de Engenharia Eletrênica, Universidade Federal de Minas Gerais, Escola de Engenharia, Belo Horizonte, Brasil.

Predicting stock market trends remains a heavily researched topic due to the inherent complexity of market fluctuations, driven by subjective and often unpredictable factors. These challenges make accurate predictions difficult for traditional models. Fuzzy models address this issue by accounting for the uncertainty of market returns, thereby mitigating the influence of human emotions on predictions. A key component of these models is the use of data structures to define clusters effectively. In this context, this paper introduces a novel fuzzy model that leverages the red-black tree (RBT) data structure to enhance predictive capability. As a balanced binary search tree, the RBT plays a critical role in improving forecasting precision. A comparative analysis with existing fuzzy models from the literature demonstrates the superior performance of the proposed approach.

Finance, Fuzzy, Statistics, Red-Black Tree, Machine Learning, Deep Learning

Tavares, T. , Ferreira, B. and Mendes, E. (2025) Software for a Novel Fuzzy Time Series Forecasting Model Based on Balanced Tree Data Structures. Journal of Computer and Communications, 13, 13-28. doi: 10.4236/jcc.2025.135002.