Implementação de métodos iterativos e computacionais para resolução de sistemas de equações lineares : um estudo aplicado

The work main objective is to show how the implementation of iterative and computational methods for solving linear systems can contribute to motivation and improve learning of Mathematics classes for high school students. To achieve the objectives of this paper, a set of classes was delivered to fifteen students and the students answered two questionnaires, one at the beginning and another at the end of the program. A bibliographic review was done considering books, articles and past work about this subject. The theoretical background includes the history of linear systems. There was an analysis of linear systems solving methods studied in high school and other methods from Numerical Calculus. The main direct methods analyzed were the Gaussian elimination with pivoting and LU factorization. As iterative methods, Gauss-Jacobi and Gauss-Seidel were used. The conditions of convergence for the iterative methods were also analyzed. Examples were used to illustrate each method of resolution that is part of this paper. Many methods of computational solutions for linear systems were applied using the following software: GeoGebra, Winplot, Maxima and SciLab. All the software is freeware and open-source. The motivation for field research was the applicability of linear systems with dimension n, (considering n> 3), in various areas of knowledge. Students from public and private school system of Distrito Federal were selected as research subject. Those students attended classes, and used the methods discussed in this work, specially the iterative and computational methods. After the analysis of the classes and feedback questionnaires about the outlined methods, especially the computational methods for applied exercises, we inferred that they motivate the students to have improved learning of Mathematic related to linear systems.