Sistema de retroalimentación inteligente basado en conocimiento común para la enseñanza virtual tutorizada

Resumen

Introduction: Audience Response Systems (ARS) is a technology that allows a group of students to respond electronically to a questionnaire. This resource makes it possible to: (1) promote active learning; (2) improve motivation; (3) create an environment for shy students; (4) monitor class attendance; (5) increase collaboration between teacher and student and even among students themselves; and (6) conduct student evaluations. ARS has in recent years incorporated gamification to create a more exciting and entertaining experience for attendees. One of the applications of artificial intelligence in education is the use of neural networks or clustering systems to incorporate cognitive diagnosis techniques. These systems are trained using answers to questionnaires containing multiple-choice questions, with the aim of clustering knowledge based on these answers. The teacher provides feedback in the form of text or images for each knowledge cluster formed. In subsequent tests using the questionnaire, the system classifies students into one of the knowledge groups each time they answer the questions, providing them with personalised and instantaneous feedback associated with that specific knowledge group. Hypothesis: The hypothesis of this thesis is "if self-assessment processes with learner-centred learning were used in conjunction with gamified and intelligent learning tools to support these processes, then improvements in students' academic performance would be achieved". Objectives: The aim of this doctoral thesis is to define self-assessment processes with learner-centred learning together with gamified, personalised and intelligent self-assessment tools and their validation in the context of the European Higher Education Area. The main objective is divided into the following specific objectives: 1. Identify and study the alternative gamification techniques used in teaching, both face-to-face and virtual. 2. To study the existing literature on ARS and intelligent tutoring systems based on knowledge groups. 3. To define learner-centred self-assessment processes supported by gamified and intelligent self-assessment systems. 4. To provide gamified and intelligent self-assessment applications for tutored e-learning. 5. To evaluate the self-assessment processes and the previous applications built by means of empirical validation through experiments in a university teaching environment, in order to contrast the training effectiveness of the proposal. Methodology: A bibliometric study was used, and the state of the problem was reviewed following systematic literature review protocols that allow the study to be reproducible. In addition, an integrated Action Research method was used for the research tasks, a methodology that provides rigour and a systematic approach, including research cycles and problem-solving cycles. This methodology involves a continuous cycle of reflection, action and evaluation, with the aim of generating practical knowledge and achieving concrete improvements in a given situation or context. Statistical analysis was used to formalise the results of the experiments. Conclusions: This doctoral thesis showed evidence that the use of gamified ARS and cognitive diagnostic-based feedback systems, guided by an adequate self-assessment process, improve students' academic performance, although the results might not be extrapolable to all disciplines. Incorporating gamification elements into ARS has a different impact on student learning depending on the gamification element used. Gamification elements that provide positive reinforcement for students, such as badges, produce greater alterations in heart rate than those that are neutral, such as ranking. The type of feedback in intelligent tutoring systems based on knowledge groups is crucial to enhance their effectiveness in student learning. Educators should consider the context of application, analyse the advantages and disadvantages of gamified ARS and intelligent feedback systems in curriculum design, adopt a reflective approach and consider the overall learning objectives and types of learners.