Digital analysis of discrete fractional order cancer model by artificial intelligence

Abstract

This article presents the approximate numerical solutions of the novel constructed Difference Caputo Order Cancer (DCOC) model utilizing the iterative numerical method combined with the neural network approach. The DCOC model is systemized for five variables and is further classified into four stages for investigation. Various fractional orders are considered to study the different scenarios of the DCOC model. The iterative systems of the DCOC model have been formulated by the iterative numerical method with neural networks associated with the computational operation of the Bayesian Regularization (BR), referred to as ENNs-BR. All the results of the DCOC model are compared by their performance. For the statistical study, the data of the model considered is partitioned into 75% and 25% in two parts. Finally, the results for performance, training state, error histogram, regression, and fit are illustrated graphically.

Keywords

• Discrete fractional order Caputo Cancer model
• Iterative numerical method
• Neural networks
• Bayesian regularization
• Training state