Um modelo fuzzy para identificação de dificuldades matemáticas

Abstract

This dissertation presents the development of a fuzzy model designed to identify mathematical learning difficulties among high school students. Grounded in fuzzy rule-based systems (FRBS), the model integrates cognitive variables and aspects related to student engagement, enabling the representation of performance levels in a gradual and interpretable manner. Its hierarchical architecture is composed of three subsystems that assess engagement, performance consistency, and specific types of mathematical difficulties. The membership functions, defined using triangular and trapezoidal models, as well as the rule bases, were constructed according to pedagogical relationships involving participation, conceptual understanding, procedural skills, and problem-statement interpretation. The system was implemented in MATLAB and tested through simulated scenarios, allowing the analysis of its behavior under different combinations of student performance. This work contributes to the development of flexible and interpretable tools capable of supporting educational practices and strengthening the connection between fuzzy modeling techniques and learning assessment processes.