Uma infra-estrutura computacional para o gerenciamento de programas de ensino individualizados

Abstract

Difficulties in learning how to read and write might bring serious jeopardy to students progression in school. Researches focusing teaching methods for individualized application, based on the paradigm of stimulus equivalence, show that it is possible to significantly improve learning of reading and other skills Although they present good results, these methods have limited range, because the presence of a tutor is not always possible or economically viable. Besides, authoring and management of learning programs based on these methods are complex and demand training of tutors and specialists. Some results have already been reached concerning to informatization, though the solutions presented offer little flexibility, since they focus on teaching topics that are either too specific or too broad, which requires complementary effort for its effective usage. This dissertation explores and evaluates software engineering and distributed computing techniques in an attempt to find a solution that might diminish costs involved in the creation, application and management of individualized programs, besides allowing large scale diffusion of teaching through these methods. A software infrastructure that allows these objectives is proposed, composed by an architecture, modules to perform specific tasks, and recommendations on some patterns of architecture, project and interface in order to help the development or adaptation of new modules. In order to maximize the installed base of the developed modules and also to benefit a larger number of students, the solution promotes interoperability of Web applications, mobile devices and interactive digital television. Aiming the validation of the proposed infrastructure and also to offer an operational codification to the problem of computerized individualized learning, a modular system named GEIC was developed and deployed, being used by researchers of several universities. Despite the initial motivation of this work to offer an infrastructure for the creation of systems specific for teaching reading and writing, the solution proposed here is very broad in scope and supports other domains and learning paradigms.