Sistemas multiagentes na avaliação pedagógica e na detecção do perfil cognitivo dos alunos: proposta de modelo para uso em ambientes de ensino a distância / Multi-agent systems in learning evaluation and in detection of students learning styles: a proposed model for use in distance education environments.

Lissandra Luvizão Lazzarotto

The computer science is a great allied in the education process. Especially in distance education environment (DEE), the spread of new information and communication technologies have been critical to the broad expansion of this mode of education. These technologies include those in the field of artificial intelligence which have great capacity to adapt to the context and personalization of the environment in accordance with the characteristics of users, increasing of degree of interactivity between the environment and its users. In the field of artificial intelligence, the multi-agent systems (MAS) are important. The MAS are organized groups of autonomous agents, each agent represents a computer system, working primarily in solving of large and complex problems. Assuming that a MAS can increase the efficiency of the teaching-learning in DEE, we develop a proposal of model of a society of autonomous agents that, by focusing on the evaluation and the detection of students learning styles, helps the teacher in the definition of more appropriate procedures for improving this process. In methodological terms, the research was conducted mainly with analysis of bibliographic studies related to the topics covered in this work: education and MAS. Based on these analysis and generating basic subsidies for the construction of the proposed MAS to be used in distance education virtual environments (DEVEs), we initially designed two proposals for pedagogical models that serve to a discipline or a course full developed in these environments. In these models were introduced important aspects of the learning evaluation through the identification of the students learning styles. For this, we considered the various theoretical dimensions of these styles. From the models and using language modeling multi-agent systems (MAS-ML), we constructed a proposed MAS which includes structural and dynamic aspects. About the conclusions, we can emphasize three: 1) the models proposed in this study tend to be very complex because, among other things, involve large numbers of users and structures to be modeled; 2) the definition of various agents showed the practical possibility of constructing a model of MAS that, through autonomy and intelligence, have mechanisms to make practicable the pedagogical models proposed; and 3) the modeling of MAS can be very useful especially for DEVEs because the peculiarities surrounding the teaching-learning in these environments.

AUTOR(ES)

FONTE

DATA DE PUBLICAÇÃO

RESUMO

ASSUNTO(S)

ACESSO AO ARTIGO

Documentos Relacionados