GRIAL resources

Permanent URI for this communityhttps://repositorio.grial.eu/handle/123456789/1

Browse

Now showing 1 - 9 of 9

Asistentes de aprendizaje basados en inteligencia artificial: Principios de seguridad y experiencias de implementación en educación superior
(Dykinson, 2024-12-30) Casañ, M. J.; Alier, M.; Pereira, J.; García-Peñalvo, F. J.
El capítulo presenta el impacto y las aplicaciones de la Inteligencia Artificial Generativa (IAGen) en educación superior, centrándose en principios de seguridad y experiencias prácticas. Desde finales de 2022, herramientas como ChatGPT y Dall-E han revolucionado los métodos de enseñanza, promoviendo la personalización del aprendizaje y la automatización de procesos educativos. Sin embargo, estas tecnologías también plantean desafíos, como la privacidad de datos, las "alucinaciones" en las respuestas de los modelos, los sesgos inherentes y la dependencia tecnológica. Para garantizar una implementación segura y ética de la IAGen, los autores proponen siete principios clave: confidencialidad, alineación con estrategias educativas, prácticas didácticas, precisión, comprensión, supervisión humana y entrenamiento ético. Estos principios buscan integrar herramientas de IA de manera alineada con los valores institucionales y las normativas de privacidad. El capítulo también introduce LAMB (Learning Assistant Manager and Builder), un marco de software diseñado para crear asistentes de aprendizaje seguros y personalizados. Estos asistentes, interoperables con sistemas como Moodle, emplean recuperación aumentada por generación (RAG) para combinar datos específicos con la capacidad de los modelos de lenguaje. Un ejemplo práctico de LAMB se ilustra en un curso de negocios donde se utilizó un asistente para realizar análisis PESTLE y DAFO, mostrando una recepción positiva por parte de los estudiantes. Finalmente, se concluye que integrar la IAGen en la educación no solo debe enfocarse en su potencial innovador, sino en asegurar una aplicación ética y responsable, alineada con los objetivos educativos. Herramientas como LAMB ejemplifican cómo la IA puede ser una pieza valiosa y segura en los ecosistemas educativos.
Evaluación de la importancia de la ética, privacidad y seguridad en los estudios de Learning Analytics, en el marco de las conferencias LAK
(Servicio de Publicaciones Universidad de Zaragoza, 2019-10-09) Amo, D.; Alier, M.; Fonseca, D.; García-Peñalvo, F. J.; Casañ, M. J.; Navarro, J.
Los orígenes del Learning Analytics son difusos y rodeados por un intenso debate sobre su definición. Este debate se sitúa en el ámbito académico y comunidad científica. Además, se pretende identificar el alcance, retos y barreras en la aplicación de análisis de datos educativos. El presente artículo corresponde a una investigación más amplia, centrada en el tratamiento y gestión de datos educativos. Hemos conducido una Literature Review aplicando Text Analytics sobre los títulos, resúmenes y autores de los artículos publicados en todas las conferencias LAK (Learning Analytics & Knowledge, periodo 2011-2019). El objetivo del análisis de texto es doble. Por una parte, averiguar si en los congresos LAK existe debate alrededor de la temática “privacidad y seguridad en el tratamiento y uso datos educativos en Learning Analytics”. Por otra parte, dar una aproximación del nivel de profundidad y aportar nuevas direcciones de investigación, si así fuera necesario. El resultado refleja una amplia tendencia en los congresos LAK de hacer un tratamiento informatizado, predictivo y masivo de datos educativos para ilustrar casos de estudio, marcos teóricos y propuestas de enfoque. Muy pocos artículos presentados se concentran en ética y/o privacidad (pero con un alto impacto científico), y ninguno en seguridad.
GDPR Security and Confidentiality compliance in LMS’ a problem analysis and engineering solution proposal
(ACM, 2019-10-16) Amo, D.; Alier, M.; García-Peñalvo, F. J.; Fonseca, D.; Casany, M. J.
We have studied the main Learning Management Systems (LMSs) to comprehend how personal data is processed and stored. We found that all the users' personal information, activity, and logs are stored unencrypted on the server filesystem and databases. A user with access to such resources may have full access to all the personal information and meta-information stored. Therefore, the LMSs are very vulnerable to information leaks in front of targeted hacker attacks due to weak GDPR compliance. In this paper, we analyze this problem from a technical and operational perspective for the open-source market leader LMS Moodle, and we propose a solution and a prototype of implementation.
Generative Artificial Intelligence in Education: From Deceptive to Disruptive
(Universidad Internacional de la Rioja, 2024-03-12) Alier, M.; García-Peñalvo, F. J.; Camba, J. D.
Generative Artificial Intelligence (GenAI) has emerged as a promising technology that can create original content, such as text, images, and sound. The use of GenAI in educational settings is becoming increasingly popular and offers a range of opportunities and challenges. This special issue explores the management and integration of GenAI in educational settings, including the ethical considerations, best practices, and opportunities. The potential of GenAI in education is vast. By using algorithms and data, GenAI can create original content that can be used to augment traditional teaching methods, creating a more interactive and personalized learning experience. In addition, GenAI can be utilized as an assessment tool and for providing feedback to students using generated content. For instance, it can be used to create custom quizzes, generate essay prompts, or even grade essays. The use of GenAI as an assessment tool can reduce the workload of teachers and help students receive prompt feedback on their work. Incorporating GenAI in educational settings also poses challenges related to academic integrity. With availability of GenAI models, students can use them to study or complete their homework assignments, which can raise concerns about the authenticity and authorship of the delivered work. Therefore, it is important to ensure that academic standards are maintained, and the originality of the student's work is preserved. This issue highlights the need for implementing ethical practices in the use of GenAI models and ensuring that the technology is used to support and not replace the student's learning experience.
Learning Analytics to Assess Students’ Behavior With Scratch Through Clickstream
(CEUR-WS.org, 2018-08-31) Amo, D.; Alier, M.; García-Peñalvo, F. J.; Fonseca, D.; Casañ, M. J.
The construction of knowledge through computational practice requires to teachers a substantial amount of time and effort to evaluate programming skills, to understand and to glimpse the evolution of the students and finally to state a quantitative judgment in learning assessment. This suposes a huge problem of time and no adecuate intime feedback to students while practicing programming activities. The field of learning analytics has been a common practice in research since last years due their great possibilities in terms of learning improvement. Such possibilities can be a strong positive contribution in the field of computational practice such as programming. In this work we attempt to use learning analytics to ensure intime and quality feedback through the analysis of students behavior in programming practice. Hence, in order to help teachers in their assessments we propose a solution to categorize and understand students’ behavior in programming activities using business technics such as web clickstream. Clickstream is a technique that consists in the collection and analysis of data generated by users. We applied it in learning programming environments to study students behavior to enhance students learning and programming skills. The results of the work support this business technique as useful and adequate in programming practice. The main finding shows a first taxonomy of programming behaviors that can easily be used in a classroom. This will help teachers to understand how students behave in their practice and consequently enhance assessment and students’ following-up to avoid examination failures.
Personal Data Broker Instead of Blockchain for Students’ Data Privacy Assurance
(Springer, 2019-04-01) Amo, D.; Fonseca, D.; Alier, M.; García-Peñalvo, F. J.; Casañ, M. J.
Data logs about learning activities are being recorded at a growing pace due to the adoption and evolution of educational technologies (Edtech). Data analytics has entered the ﬁeld of education under the name of learning analytics. Data analytics can provide insights that can be used to enhance learning activities for educational stakeholders, as well as helping online learning applications providers to enhance their services. However, despite the goodwill in the use of Edtech, some service providers use it as a means to collect private data about the students for their own interests and beneﬁts. This is showcased in recent cases seen in media of bad use of students’ personal information. This growth in cases is due to the recent tightening in data privacy regulations, especially in the EU. The students or their parents should be the owners of the information about them and their learning activities online. Thus they should have the right tools to control how their information is accessed and for what purposes. Currently, there is no technological solution to prevent leaks or the misuse of data about the students or their activity. It seems appropriate to try to solve it from an automation technology perspective. In this paper, we consider the use of Blockchain technologies as a possible basis for a solution to this problem. Our analysis indicates that the Blockchain is not a suitable solu-tion. Finally, we propose a cloud-based solution with a central personal point of management that we have called Personal Data Broker.
Privacidad, seguridad y legalidad en soluciones educativas basadas en Blockchain: Una Revisión Sistemática de la Literatura
(2020-05-06) Amo Filvà, D.; Alier, M.; García-Peñalvo, F. J.; Fonseca, D.; Casañ, M. J.
La Analítica del Aprendizaje (proveniente del término en inglés Learning Analytics) procesa los datos de los estudiantes, incluso los estudiantes menores de edad. El ciclo analítico consiste en recoger datos, almacenarlos durante largos períodos y utilizarlos para realizar análisis y visualizaciones. A mayor cantidad de datos, mejores resultados en el análisis. Este análisis puede ser descriptivo, predictivo e, incluso, prescriptivo, lo que implica la gestión, el tratamiento y la utilización de datos personales. El contexto educativo es, por lo tanto, muy sensible, a diferencia de los contextos individuales en los que el análisis se utiliza a voluntad. No está claro cómo están utilizando los datos de los estudiantes las empresas de tecnología que dan servicio en educación y a quiénes realmente se les beneficia, cómo esto afectará a los estudiantes en un futuro a corto y largo plazo, o qué nivel de privacidad o seguridad se aplica para proteger los datos de los estudiantes. Por consiguiente, y en relación con lo expuesto, el análisis de datos educativos implica un contexto sensible y de fragilidad en la gestión y análisis de datos personales de los estudiantes, incluidos menores, en el que hay que maximizar las precauciones. En esta revisión sistemática de la literatura se explora la importancia de la protección y seguridad de los datos personales en el campo de la educación mediante las promesas emergentes de los interesados en usar la tecnología blockchain. Los resultados denotan que es importante entender las implicaciones y riesgos derivados de usar tecnologías emergentes en educación, su relación con la sociedad y la legalidad vigente
Protected Users: A Moodle Plugin To Improve Confidentiality and Privacy Support through User Aliases
(MDPI, 2020-03-24) Amo, D.; Alier, M.; García-Peñalvo, F. J.; Fonseca, D.; Casañ, M. J.
The privacy policies, terms, and conditions of use in any Learning Management System (LMS) are one-way contracts. The institution imposes clauses that the student can accept or decline. Students, once they accept conditions, should be able to exercise the rights granted by the General Data Protection Regulation (GDPR). However, students cannot object to data processing and public profiling because it would be conceived as an impediment to teachers to execute their work with normality. Nonetheless, regarding GDPR and consulted legal advisors, a student could claim identity anonymization in the LMS, if adequate personal justifications are provided. Per contra, the current LMSs do not have any functionality that enables identity anonymization. This is a big problem that generates undesired situations which urgently requires a definitive solution. In this work, we surveyed students and teachers to validate the feasibility and acceptance of using aliases to anonymize their identity in LMSs as a sustainable solution to the problem. Considering the positive results, we developed a user-friendly plugin for Moodle that enables students' identity anonymization by the use of aliases. This plugin, presented in this work and named Protected users, is publicly available online at GitHub and published under GNU General Public License.
Unplugged institutions: towards a localization of the cloud for Learning Analytics privacy enhancement
(CEUR-WS.org, 2022-10-09) Amo-Filvà, D.; Fonseca, D.; Alier, M.; García-Peñalvo, F. J.; Casañ, M. J.
The debate on privacy issues in Learning Analytics processes has been going on for a long time. In academic terms, various researchers attempted to identify the origin of the problem, provide solutions, and propose alternatives. However, the problem is complex, not yet solved, and increasingly pressing and serious. We reflect on cloud computing technologies as a generator of privacy issues and new derivatives. We assume that the technology used in the cloud is aggravating the problem, not Learning Analytics itself. Considering data capitalism, we argue that it is hopelessly impossible to solve the privacy problem, nor even mitigate it, when educational institutions use data ubiquity services in the cloud. We point to the paradox of Learning Analytics as the in-compatibility factor with third-party cloud computing services, where the latter is the link to all the associated privacy issues. To mitigate privacy issues, we propose the deconstruction of cloud computing for its localization. The localization is the basis of a new concept related to the disconnection of educational institutions from the cloud. New technological perspectives, legal frameworks, and social, cultural, and political changes are required.