Programas de cómputo para la generación automática de exámenes matemáticos y la evaluación en el entorno de Internet

Abstract

Se describe la arquitectura, pruebas resultados de tres programas desarrollados para sistemas de aprendizaje "virtual", dos de ello realizan evaluación automatizada, denominados programas de evaluación "clásico" y "modificado". Un tercer programa corresponde a un generador aleatorio de exámenes matemáticos.

En el presente documento se expone la estructura del proyecto de tesis, los procesos investigación realizados para su conclusión, así como el marco conceptual que la sustenta. Se ha dado importancia a la modelación de los programas de cómputo propuestos, tanto de la arquitectura, como de las funciones desarrolladas; todo ello en términos de UML (Unified Modeling Language, Lenguaje de Modelado Unificado). Tales actividades no sólo permiten entender el software, sino también facilitar su posterior evolución y adecuación o inserción en un sistema de evaluación de mayor alcance y versatilidad.

Se exponen los resultados de las pruebas realizadas a los programas desarrollados en su ejecución frente a diferentes navegadores, así como la comparativa con desarrollos no nacionales que se anuncian como Generadores de Exámenes. // The product of this work are three computer programs: Two are autoevaluation and third one it is a generating program of mathematical examinations of random form, all of them applicable to the Internet environment.

In the thesis an historical and labor research has been realized, for to be situed the field of computational applications of proposed programs and education modality in the one that is inserted.

The problem of the evaluation treats itself in the education and the characteristic of the reactives of pedagogical proofs; which must be considered on having for developed good reactives in the systems of automated evaluation “Online”.

One exams several case of computer programs of management knowledge of national origin as foreign as Moodle, BlackBoard, EVA, SOFIA; emphasizing his capacities of automated evaluations and his modalities of pedagogical proofs. Also checks tools for elaboration of reactives of pedagogical proofs, KnowledegePresenter of Flash between them.

The review, not only contributed to characterize the current situation of these tools and programs of management learning, but also to establish the limitations of the national systems of evaluation, restricted often to test of multiple option and evaluation of the learnings.

The computer programs developed, are good offer for the case of evaluation of the learning; them contributing to the possibility of feed-back for the pupil and not only to deliver him, an accounting mistake. Another contribution is it of the generation of exams allows multimedia. However, must the programs are incorporate to “robust” systems of evaluation, in which this ones the evaluation not only is direct on the student, also too the teacher, authority, institute and society, linking the results to management Data Base.

In this document explain the structure of the thesis project, research process to realized, also the conceptual frame that sustains her. One has given importance to the modeling of computer programs proposed, so much the architecture, so development function, all in UML's (Unified Modeling Language) terms. Such activities not only allow understand the software, but too facilitate his later evolution and adequacy or insertion in a system of evaluation and adequacy in a system of evaluation of major scope und versatility.