Francisco J. Gallego-Durán
University of Alicante
Network
Latest external collaboration on country level. Dive into details by clicking on the dots.
Publication
Featured researches published by Francisco J. Gallego-Durán.
Revista Iberoamericana De Tecnologías Del Aprendizaje | 2016
Faraón Llorens-Largo; Francisco J. Gallego-Durán; Carlos Villagrá-Arnedo; Patricia Compañ-Rosique; Rosana Satorre-Cuerda; Rafael Molina-Carmona
Although several definitions of gamification can be found in the literature, they all have in common certain aspects: the application of strategies, models, dynamics, mechanics, and elements of the games in other contexts than games, and the objective of producing a playful experience that fosters motivation, involvement, and fun. In this paper, our approach gamifying the learning process of a subject is presented. Our experience throughout time in using games and gamification in learning have led us to propose, lately, a personalized, automated, and gamified learning system. As a result of this experience and after several years of continuous feedback from our students, we have learned several lessons on how to approach the task of gamification. These lessons are summarized in the following concepts: fun, motivation, autonomy, progressiveness, feedback, error tolerance, experimentation, creativity, and adaptation to the specific case. The final aim is sharing our experience and opening a debate about what key elements the gamification lie in.
EC-TEL | 2015
Carlos Villagrá-Arnedo; Francisco J. Gallego-Durán; Rafael Molina-Carmona; Faraón Llorens-Largo
A prediction system to early detect learning problems is presented. The starting point is a gamified learning system from which a massive set of usage and learning data is collected. They are analyzed using Machine Learning techniques and a prediction of each student’s performance is obtained. The information is weekly presented as a progression chart, with valuable information about students’ progression. The system has a high degree of automation, is progressive, uses learning outcomes as well as usage data, allows the evaluation and prediction of the acquired skills, and contributes to a truly formative assessment.
international conference on learning and collaboration technologies | 2016
Carlos Villagrá-Arnedo; Francisco J. Gallego-Durán; Rafael Molina-Carmona; Faraón Llorens-Largo
Gamification is set to be a disruptive innovation in the field of education in the next years, as a way to encourage learning, since when the fun impregnates the learning process, motivation increases and stress is reduced. However, most experiences in learning gamification just remain on the surface, just offering a layer of standardized game elements such as badges, leader boards and medals. Instead, a deeper transformation of the learning process is needed, making up a true process reengineering. As a practical example, PLMan learning system is presented, an attempt to redefine the learning process in the context of a particular subject. It is based on a unique and simple type of problem: solving mazes of the PLMan game, an adaptation of the famous Pac-Man game. The maze, as the building block of our learning strategy, has a set of properties that allows us to introduce all the features of games in the learning process. From this experience some important lessons about the gamification of the teaching-learning process can be obtained: the importance of fun as a consequence of learning, the need of having an immediate feedback of our actions, the trial and error possibility as a major source of learning and progress, the relevance of experimentation and creativity as a means to develop the learners skills and the importance of autonomy to give the learners the control of their learning process. All these features are propellants for learning and a way to improve the motivation of learners.
Universal Access in The Information Society | 2018
Francisco J. Gallego-Durán; Rafael Molina-Carmona; Faraón Llorens-Largo
An effective adaptive learning system would theoretically maintain learners in a permanent state of flow. In this state, learners are completely focused on activities. To attain this state, the difficulty of learning activities must match learners’ skills. To perform this matching, it is essential to define, measure and deeply analyze difficulty. However, very few previous works deal with difficulty in depth. Most commonly, difficulty is defined as a one-dimensional value. This permits ordering activities, but limits the possibilities of deep analysis of activities and learners’ performance. This work proposes a new definition of difficulty and a way to measure it. The proposed definition depends on learners’ progress on activities over time. This expands the concept of difficulty over a two-dimensional space, also making it drawable. The difficulty graphs provide a rich interpretation with insights into the learning process. A practical case is presented: the PLMan learning system. This system is formed by a web application and a game to teach computational logic. The proposed definition is applied in this context. Measures are taken and analyzed using difficulty graphs. Some examples of these analyses are shown to illustrate the benefits of this proposal. Singularities and interesting spots are easily identified in graphs, providing insights in the activities. This new information lets experts adapt the learning system by improving activity classification and assignment. This first step lays solid foundations for automation, making the PLMan learning system fully adaptive.
International Journal of Design & Nature and Ecodynamics | 2016
Carlos Villagrá-Arnedo; Francisco J. Gallego-Durán; Patricia Compañ-Rosique; Faraón Llorens-Largo; Rafael Molina-Carmona
The volume and quality of data, but also their relevance, are crucial when performing data analysis. In this paper, a study of the influence of different types of data is presented, particularly in the context of educational data obtained from Learning Management Systems (LMSs). These systems provide a large amount of data from the student activity but they usually do not describe the results of the learning process, i.e., they describe the behaviour but not the learning results. The starting hypothesis states that complementing behavioural data with other more relevant data (regarding learning outcomes) can lead to a better analysis of the learning process, that is, in particular it is possible to early predict the student final performance. A learning platform has been spe cially developed to collect data not just from the usage but also related to the way students learn and progress in training activities. Data of both types are used to build a progressive predictive system for helping in the learning process. This model is based on a classifier that uses the Support Vector Machine technique. As a result, the system obtains a weekly classification of each student as the probability of belonging to one of three classes: high, medium and low performance. The results show that, supplementing behavioural data with learning data allows us to obtain better predictions about the results of the students in a learning system. Moreover, it can be deduced that the use of heterogeneous data enriches the final performance of the prediction algorithms.
international conference on learning and collaboration technologies | 2016
Francisco J. Gallego-Durán; Rafael Molina-Carmona; Faraón Llorens-Largo
In any learning environment, training activities are the basis for learning. Students need to practice to develop new skills and improve previously acquired abilities. Each student has specific needs based on their previous knowledge and personal skills. The allocation of a proper activity for a particular student consists in selecting a training activity that fits the skills and knowledge of the student. This allocation is particularly important since students who are assigned a too hard training activity will tend to leave it rather than making the necessary effort to complete it. Moreover, when the activity is too easy it does not represent a challenge for the student and the learning outcomes will tend to be very limited. An motivating activity, suitable for a given student, should be neither too easy nor too difficult. The problem arises when trying to measure or estimate the difficulty given any training activity. Our proposal is a definition of difficulty of a learning activity that can be used to measure the learning cost of a general learner. As a first step, the desirable features and the intrinsic limitations of a difficulty function are identified, so that a mathematical definition can be obtained quite straightforward. The result is an intuitive, understandable and objectively measurable way to determine the difficulty of a learning activity.
international conference on human-computer interaction | 2018
Francisco J. Gallego-Durán; Carlos Villagrá-Arnedo; Rosana Satorre-Cuerda; Patricia Compañ-Rosique; Faraón Llorens-Largo
There are subjects in which teaching and learning is hard by experience. Some subjects in physics, maths or computing seem to be difficult by nature. Teachers test many ways to help student learn these subjects. In Computer Programming the approach seems to be using higher-level languages, concepts and abstractions. It seems reasonable that languages similar to human language can ease the task of computer programming. Similar ideas are explored in other subjects. However, this seems contradictory with the way we construct knowledge: lower-level concepts support the development of higher-level ones. Is it possible to master higher-level concepts without previously mastering lower-level ones?
technological ecosystems for enhancing multiculturality | 2017
Rafael Molina-Carmona; Carlos Villagrá-Arnedo; Francisco J. Gallego-Durán; Faraón Llorens-Largo
Learning Analytics1 is a powerful tool that provides rich information for students, teachers and academic authorities. There is a wide range of possible applications, and one of them is leveraging the information to improve the instructional design of a course. In this research, we introduce the results of a Learning Analytics engine to improve all the stages of an Action Research experience. We have carried out three iterations: iteration 1, devoted to design an instructional course using an automated learning platform that collects data from the students; iteration 2, focused on the analysis of the individual and aggregated data collected from the students to obtain group behaviors; and iteration 3, currently under way, devoted to improve the structure of the course using the results of the previous iterations. We have made use of some graphical representations of the data that help to understand the aggregated data and to detect important events and moments of intervention. We have detected that the behavior of the students is strongly conditioned by the deadline structure of the course and that there is usually a crucial moment, by halfway of the course, where the situation is tending to stabilize and which is a good moment to reinforce the supervision on the student. Finally, we have stated that low performance students are usually recoverable to the last moment.
international conference on learning and collaboration technologies | 2017
Vicente A. Quesada Mora; Francisco J. Gallego-Durán; Rafael Molina-Carmona; Faraón Llorens-Largo
In video games, organic tutorials are first levels of the games, designed to teach their basic controls while the player plays. They provide some kind of subliminal learning, are very effective and natural and teach without losing the fun, but they are not easy to be properly designed. The purpose of this research is assisting the designers in the task of defining organic tutorials by proposing a guide of design principles and patterns. After reviewing the learning theories and design techniques for organic tutorials, an analysis of some representative video games is performed. Then, a guide made of ten principles is proposed. This guide helps the developers to clearly understand the fundamentals of organic tutorials and sheds light on what games teach us. It helps to understand some kind of subliminal learning and opens the way to design other learning experiences based on the proposed principles.
Conference of the Spanish Association for Artificial Intelligence | 2013
Francisco J. Gallego-Durán; Rafael Molina-Carmona; Faraón Llorens-Largo
Neuroevolution has come a long way over the last decade. Lots of interesting and successful new methods and algorithms have been presented, with great improvements that make the field become very promising. Concretely, HyperNEAT has shown a great potential for evolving large scale neural networks, by discovering geometric regularities, thus being suitable for evolving complex controllers. However, once training phase has finished, evolved neural networks stay fixed and learning/adaptation does not happen anymore. A few methods have been proposed to address this concern, mainly using Hebbian plasticity and/or Compositional Pattern Producing Networks (CPPNs) like in Adaptive HyperNEAT. This methods have been tested in simple environments to isolate the effectiveness of adaptation from the Neuroevolution. In spite of this being quite convenient, more research is needed to better understand online adaptation in more complex environments. This paper shows a new proposal for online weight adaptation in neuroevolved artificial neural networks, and presents the results of several experiments carried out in a race simulation environment.