Generación automática de preguntas cloze para cuestionarios Moodle sobre análisis léxico

  1. Roberto Izquierdo Amo
  2. Pedro Latorre Carmona
  3. César I. García Osorio
  4. José A. Barbero Aparicio
Journal:
Actas de las Jornadas sobre la Enseñanza Universitaria de la Informática (JENUI)

ISSN: 2531-0607

Year of publication: 2021

Issue: 6

Pages: 163-170

Type: Article

More publications in: Actas de las Jornadas sobre la Enseñanza Universitaria de la Informática (JENUI)

Abstract

Moodle questionnaires are a convenient way to face the online assessment, so relevant in these pandemic ti- mes. Unfortunately, their preparation is slow, cumber- some, and prone to mistakes. In addition to designing the question, then it has to be introduced in Moodle. If the question has images, it is necessary to create and upload them to Moodle. It is also necessary to configu- re the grades associated with the correct answer and the incorrect alternative options. If this process has to be repeated with many questions, the time required ends up being prohibitive. In this work, a tool is presented that solves all these problems, allowing the automatic and massive generation of questions for the evaluation of algorithms for obtaining finite automata from regu- lar expressions. The evaluation of these algorithms is relevant in both formal languages and compiler design subjects. With a genetic algorithm, exercises of confi- gurable complexity are searched, after which the text, tables and associated images are generated for the mas- sive loading of all the questions generated. The main impact of the use of this tool is in the enormous reduc- tion of the time in the preparation of questions, repla- cing hundreds or thousands of interactions with Mood- le forms, for a few steps.