Examinando por Autor "Palacios-Alonso, Daniel"

Mostrando 1 - 8 de 8

A guided Scratch visual execution environment to introduce programming concepts to CS1 students
(MDPI, 2021-09-17) Hijón-Neira, Raquel; Connolly, Cornelia; Palacios-Alonso, Daniel; Borrás-Gené, Oriol
First-year computer science (CS1) university students traditionally have difficulties understanding how to program. This paper describes research introducing CS1 students to programming concepts using a Scratch programming language guided visual execution environment (VEE). The concepts addressed are those from an introductory programming course (sequences, variables, operators, conditionals, loops, and events and parallelism). The VEE guides novice students through programming concepts, explaining and guiding interactive exercises executed in Scratch by using metaphors and serious games. The objective of this study is, firstly, to investigate if a cohort of 124 CS1 students, from three distinct groups, studying at the same university, are able to improve their programming skills guided by the VEE. Secondly, is the improvement different for various programming concepts? All the CS1 students were taught the module by the same tutor in four 2-h sessions (8 h), and a qualitative research approach was adopted. The results show students significantly improved their programming knowledge, and this improvement is significant for all the programming concepts, although greater for certain concepts such as operators, conditionals, and loops than others. It also shows that students lacked initial knowledge of events and parallelism, though most had used Scratch during their high school years. The sequence concept was the most popular concept known to them. A collateral finding in this study is how the students’ previous knowledge and learning gaps affected grades they required to access and begin study at the university level.
Automatic Competitions in the Unibotics open online robot programming web
(Springer, 2022-11-19) Fernández-Ruiz, Raúl; Palacios-Alonso, Daniel; Cañas-Plaza, J.M.; Roldán-Álvarez, David
This paper presents a gamification extension for the Uni botics educational robotics web platform, which provides the ability to organize online international competitions about robot programming. Unibotics includes several robotics challenges, the robots may be pro grammed in Python from the browser. It is ROS-based, cross-platform and uses Gazebo robotics simulator. The competition process has been fully automated creating an auxiliary user, named CeremonyMaster, which may load and run the submitted code from all the participants, and evaluate them providing a score depending on their performances when solving the challenge. At the same time the live results can be seen at a dynamic web page and the whole unfolding of the tournament is broadcasted through Twitch. The underlying goal of this extension is to increase the motivation and learning engagement of university students in computer science or robotics engineering areas. The extension has been experimentally validated with a first competition and more than 50 university students. The proposed challenge was to program a Formula 1 car, endowed with a camera, to follow a red line along a race circuit, the faster the better.
Computational approaches to Explainable Artificial Intelligence: Advances in theory, applications and trends
(Elsevier, 2023-07-19) Górriz, J.M.; Álvarez-Illán, A.; Álvarez-Marquina, Agustín; Arco, J.E.; Atzmueller, M; Ballarini, F; Barakova, Emilia; Bologna, G; Bonomini, Paula; Castellanos-Domínguez, Germán; Castillo-Barnes, D.; Cho, S.B; Contreras, Ricardo; Cuadra, J.M; Domínguez, E.; Domínguez-Mateos, Francisco; Duro, Ricardo; Elizondo, D; Fernádez-Caballero, A; Fernádez-Jover, Eduardo; Formoso, M.A.; Gallego-Molina, N.J.; Gamazo, J.; García-González, J.; García-Rodríguez, J.; Garre, Carlos; Garrigós, J.; Gómez-Rodellar, Andrés; Gómez-Vilda, Pedro; Graña, Manuel; Guerrero-Rodríguez, B.; Hendrikse, S.C.F.; Jiménez-Mesa, C.; Jodra-Chuan, Marina; Julian, V.; Kotz, G.; Kutt, K.; Leming, M.; de Lope, Javier; Macas, B.; Marrero-Aguiar, V.; Martínez, J.J.; Martinez-Murcia, F.J.; Martínez-Tomás, R.; Mekyska, Jiri; Nalepa, G.J.; Novais, P.; Orellana, D.; Ortiz, A.; Palacios-Alonso, Daniel; Palma, J.; Pereira, A.; Pinacho-Davidson, P.; Pinninghoff, María Angélica; Ponticorvo, M.; Psarrou, A.; Ramírez, Javier; Rincón, M.; Rodellar-Biarge, Victoria; Rodríguez-Rodríguez, I.; Roelofsma, P.H.M.P.; Santos, J.; Salas-Gonzalez, D.; Salcedo-Lagos, P.; Segovia, Francisco; Shoeibi, A.; Silva, M.; Simic, D.; Suckling, J.; Treur, J.; Tsanas, Athanasios; Varela, R.; Wang, S.H.; Wang, W.; Zhang, Y.D.; Zhu, H.; Zhu, Z.; Ferrández-Vicente, J.M.
Deep Learning (DL), a groundbreaking branch of Machine Learning (ML), has emerged as a driving force in both theoretical and applied Artificial Intelligence (AI). DL algorithms, rooted in complex and non-linear artificial neural systems, excel at extracting high-level features from data. DL has demonstrated humanlevel performance in real-world tasks, including clinical diagnostics, and has unlocked solutions to previously intractable problems in virtual agent design, robotics, genomics, neuroimaging, computer vision, and industrial automation. In this paper, the most relevant advances from the last few years in Artificial Intelligence (AI) and several applications to neuroscience, neuroimaging, computer vision, and robotics are presented, reviewed and discussed. In this way, we summarize the state-of-the-art in AI methods, models and applications within a collection of works presented at the 9th International Conference on the Interplay between Natural and Artificial Computation (IWINAC). The works presented in this paper are excellent examples of new scientific discoveries made in laboratories that have successfully transitioned to real-life applications.
Creating Realistic Presentation Attacks for Facial Impersonation Step-by-Step
(Editorial Institute Of Electrical And Electronics Engineers Inc., 2023) Gallardo Cava, Roberto; Ortega del Campo, David; Guillen García, Julio; Palacios-Alonso, Daniel; Conde, Cristina
Los ataques de presentación son uno de los muchos peligros a los que se enfrentan hoy en día las fuerzas del orden. Además la ciencia de los materiales avanza constantemente y los delincuentes, conscientes de este hecho, aprovechan los nuevos compuestos para fabricar nuevos artefactos que les permitan cruzar fronteras burlando los puntos de control fronterizo. Este artículo presenta la creación de varios ataques de presentación utilizando maquillaje, látex hiperrealista y máscaras protésicas. Cabe señalar que este tipo de ataques no recibe la atención adecuada, debido a la dificultad en su elaboración. Se requiere el trabajo de profesionales del sector del maquillaje. Cada etapa de la elaboración se se analiza para detectar cualquier artefacto que facilite la detección del ataque, utilizando un enfoque multiespectral enlos espectros visible y térmico. La metodología evalúa tres sistemas diferentes de reconocimiento facial (FRS), las diferentes etapas de la suplantación, es decir, cuando una parte específica de la cara como la nariz, los pómulos, la mandíbula, o los ojos. Los resultados muestran que determinadas partes de la cara mejoran la suplantación y dificultan más difícil para los algoritmos detectar una posible suplantación. Sin embargo, otras partes de la cara, como la mandíbula, no sólo no mejoran la suplantación sino que empeoran significativamente el rendimiento. Utilizando OpenFace como ejemplo de FRS, que es uno de los FRS empleados en este trabajo de investigación, la comparación de buena fe de el objetivo arroja una puntuación de 0,304, mientras que con el ataque de maquillaje antes de aplicar maquillaje a la mandíbula, da 0,291, y después de aplicar el maquillaje, da 0,421.
Deep Learning-Based Gender Classification by Training With Fake Data
(IEEE, 2023-10-27) Oulad-Kaddour, Mohamed; Haddadou, Hamid; Conde-Vilda, Cristina; Palacios-Alonso, Daniel; Benatchba, Karima; Cabello, Enrique
Gender classification of human faces is a trending topic and a remarkable biometric task. This research area has useful applications in several fields, such as automated border control (ABC) and forensic work. There are many approaches to gender classification in the literature; the classical approaches usually use real faces. Although good performances have been achieved, data collection remains a problem. Additionally, the privacy of individuals must be included in many existing works. These drawbacks can be overcome by using fake faces. Recently, the creation of a robust fake face corpus using machine learning has become possible. Our main contribution in the present paper is to experimentally investigate the ability of an artificial deepfake corpus to be a substitute for real corpora in facial gender classification tasks.We propose a deep learning-based approach using convolutional neural networks trained with fake faces and tested on real faces. By exploiting artificial faces, data collection obstacles are resolved for the training step, and privacy is highly preserved. Four classifiers based on popular convolutional neural network architectures were implemented. In the test phase, we used faces of real identities extracted from well-known experimental databases such as Face Recognition Technology (FERET), Faculdade de Engenharia Industrial (FEI) faces, Face Recognition and Artificial Vision (FRAV) and Labeled Faces in theWild (LFW). The results achieved are very promising. We obtained high accuracy rates and low EER scores. They are similar to those of research works using real faces. As a result of this work, we propose a gender-labeled deepfake facial dataset containing more than 200k deepfake corpora that we will make available upon request for research purposes.
Experiences and Proposals of Use of Generative AI in Advanced Software Courses
(IEEE Computer Society Press, 2024-07-08) Palacios-Alonso, Daniel; Urquiza-Fuentes, Jaime; Velázquez-Iturbide, J. Ángel; Guillén-García, Julio
The last year, we have witnessed the popularization of generative artificial intelligence. Its output includes text, code, image, audio, speech, voice, music, and video. Therefore, it impacts education courses where students are required to elaborate on any of these artifacts. In particular, the generation of code affects informatics courses, where assignments usually ask students to develop and deliver programming code. The impact of generative artificial intelligence on informatics courses has been mainly studied for introductory programming courses. These studies have shown that generative artificial intelligence is able to produce highly sophisticated programs, but also that its results and rationale can be inaccurate. Moreover, the impact of generative artificial intelligence has not been studied for other informatics subjects. In this paper, we present our preliminary experience and proposals on three advanced software courses, namely video games, advanced algorithms and language processors. For the video games course, we present the opportunities of use of generative artificial intelligence and the results of a survey conducted with students on their use to obtain different media products. For the algorithms course, we present the result of a session driven by the instructor on different design techniques, showing the merits and demerits of the answers generated. For the language processors course, a proposal of use of generative artificial intelligence is presented, broken down into the parts of a typical language processor. The paper concludes with some suggestions for instructors.
Identification of Smith–Magenis syndrome cases through an experimental evaluation of machine learning methods
(Frontiers Media, 2024-03-22) Fernández-Ruiz, Raúl; Núñez-Vidal, Esther; Hidalgo-deLaGuía, Irene; Garayzábal-Heinze, Elena; Álvarez-Marquina, Agustín; Martínez-Olalla, Rafael; Palacios-Alonso, Daniel
This research work introduces a novel, nonintrusive method for the automatic identification of Smith–Magenis syndrome, traditionally studied through genetic markers. The method utilizes cepstral peak prominence and various machine learning techniques, relying on a single metric computed by the research group. The performance of these techniques is evaluated across two case studies, each employing a unique data preprocessing approach. A proprietary data “windowing” technique is also developed to derive a more representative dataset. To address class imbalance in the dataset, the synthetic minority oversampling technique (SMOTE) is applied for data augmentation. The application of these preprocessing techniques has yielded promising results from a limited initial dataset. The study concludes that the k-nearest neighbors and linear discriminant analysis perform best, and that cepstral peak prominence is a promising measure for identifying Smith–Magenis syndrome.
Mejora de una asignatura para la formación del profesorado en programación basada en bloques
(Asociación de Enseñantes Universitarios de la Informática (AENUI), 2023) Velázquez-Iturbide, J. Ángel; Paredes-Velasco, Maximiliano; Cavero Díaz, Sergio; Palacios-Alonso, Daniel
Uno de los principales retos para la introducción de una materia obligatoria de informática en niveles educativos preuniversitarios es la falta de profesorado formado en informática. En nuestra universidad ofrecemos un máster para formar profesores en competencia digital y programación. La asignatura “Programación y Pensamiento Computacional I” presenta una introducción a la programación basada en bloques. En el curso académico 2021/22 se realizó un diseño de la asignatura basada en cuatro lenguajes de bloques en orden creciente de complejidad. Aunque los alumnos valoraron muy positivamente la asignatura, se identificaron varias cuestiones mejorables. En la comunicación se presentan los cambios introducidos durante el curso 2022/23, que consisten en la eliminación del lenguaje Code.org, una revisión de los apuntes de Scratch, el desarrollo de nuevos ejercicios de autoestudio para ScratchJr y Scratch, y la transición de Scratch a App Inventor. Se presentan los resultados obtenidos de rendimiento de los alumnos y de aceptación de la asignatura. La asignatura ha consolidado su aceptación por los alumnos, pero los cambios introducidos no han redundado en una mejora apreciable y aún persiste como reto el aprendizaje de los elementos más complejos, principalmente App Inventor. | One of the main challenges to introduce informatics as a mandatory subject matter in pre-college education is the lack of teachers adequately trained on informatics. Our university offers master’s studies aimed at teachers’ development in digital competence and computer programming. The course “Programming and computational thinking I” introduces block-based programming. In the academic year 2021/22, the course was designed as a sequence of four languages, in increasing order of complexity. The students rated the course very high, but a few issues were amenable to improvement. In this paper, we present the changes introduced for the academic year 2022/23, comprising the removal of Code.org, re-elaboration of Scratch lecture notes, development of additional self-study exercises for ScratchJr and Scratch, and transition between Scratch and App Inventor. The paper also presents the out-comes obtained on students’ performance and course acceptance. The course is consolidated according to students’ high acceptance. However, the changes introduced did not produce a significant enhancement of acceptance, and learning the most complex elements remains an open challenge, especially App Inventor.