15879 research outputs found
Sort by
Sustainability and advanced materials
No full textEducational resource in English focusing on sustainability and advanced materials. It includes a reading on sustainable materials, market forces, and future directions such as circular economy, biomimicry, and digital simulation. Learners answer comprehension questions, practice pronunciation, complete keyword sentences, engage in speaking and watching activities with a related video, and review grammar on third and mixed conditionals, concluding with debate and Kahoot review
Novedades ES6: funciones flecha
No full textUnidad didáctica que presenta las funciones flecha de ES6 en JavaScript. Compara su sintaxis con las funciones tradicionales, muestra ejemplos de retorno implícito, interpolación de cadenas, devolución de objetos, funciones anónimas y el comportamiento de this. Incluye fragmentos de código para ilustrar usos comunes y buenas prácticas
Beyond grammar. Can AI undestand the nuanaces of academic writing?
Get PDFOn July 4, 2024, the free versions of ChatGPT, Gemini, and Claude were evaluated for their effectiveness in grading technical English assignments in an academic context. The study sought to determine ifthese AI platforms could serve as reliable grading tools by analyzing their responses to two sets of papers. The first set contained original student submissions written in technical English, while the second consisted of the same papers rewritten in grammatically correct but academically incoherent language. The AI models provided three grading suggestions for each assignment and were also asked to assign a final grade. These AI-generated grades were then compared with the grades given through traditional manual grading by educators. The analysis focused on the consistency of the AI's grading, its alignment with human assessments, and its ability to distinguish between technical precision and overall linguistic coherence. The findings contribute to ongoing discussions about the role of AI in education, particularly in the assessment of academic writing. This study highlights the potential benefits and limitations of using AI as a grading tool, emphasizing the need for further refinement before such systems can be reliably implemented in academic settings
Influence of melatonin and salinity on salicornia spp yield and shelf life
No full text[SPA] La salicornia es una planta halófita destacada por sus múltiples beneficios, tanto para el consumo humano, como sustituto a la sal convencional, como para otros muchos ámbitos, entre los más destacados, la recuperación de suelos afectados por niveles altos de salinidad. En este trabajo se ha estudiado su cultivo en condiciones salinas bajo aplicaciones de melatonina para mejorar su productividad.
En este proyecto, se evaluó el comportamiento del cultivo Salicornia spp. a diferentes niveles de salinidad (0-50-100-200 mM de NaCl) y cuatro concentraciones de melatonina exógena (0-50-100 y 200 μM) aplicadas de manera foliar. El experimento se llevó a cabo mediante un sistema de bandejas flotantes bajo condiciones de invernadero.
Tras finalizar el cultivo, se evaluaron los parámetros de rendimiento, crecimiento y comportamiento post cosecha. Los resultados indican que la alta salinidad, combinada con dosis moderadas de melatonina, desempeña un papel relevante en el rendimiento y la acumulación de biomasa, especialmente en la biomasa radicular y la altura de las plantas. La aplicación exógena de melatonina favoreció además la conservación y calidad de los brotes en post cosecha. Aunque en las condiciones estudiadas no se observó un efecto directo sobre el incremento del rendimiento, su uso junto con niveles moderados de salinidad mostró una influencia positiva al mejorar la tolerancia al estrés salino y el comportamiento post cosecha del cultivo.
[ENG] Salicornia is a halophytic plant noted for its multiple benefits, ranging from human consumption as a functional substitute for conventional salt and for several other applications, most notably the rehabilitation of soils affected by high salinity levels. To optimize its cultivation and productivity, this study investigated whether the exogenous application of melatonin could enhance growth and yield performance.
The behaviour of Salicornia spp. was studied at different levels of salinity (0-50-100- 200 mM NaCl) combined with different melatonin concentrations (0-50-100 and 200 μM) The plant completed their reproductive cycle in a floating system under greenhouse conditions.
After completing their growth cycle, the yield, growth and post-harvest behaviour parameters of Salicornia spp. were evaluated. Results indicate that high salinity, combined with moderate doses of melatonin, plays a significant role in yield and biomass accumulation, especially in root biomass and plant height. Furthermore, exogenous application of melatonin also favoured the preservation and quality of shoots after harvest. Although no direct effect on yield increase was observed under the conditions studied, the synergistic use of melatonin with moderate salinity levels positively influenced the crop by improving salt stress tolerance and post-harvest behaviour of the crop.Escuela Técnica Superior de Ingeniería AgronómicaUniversidad Politécnica de Cartagen
Contribution to the study and development of intrusion and threat detection systems in IoT
Get PDF[SPA] El crecimiento del IoT ha expandido exponencialmente la superficie de ataque generando amenazas que los IDS tradicionales no abordan eficazmente por sus limitaciones de adaptación, alto coste de mantenimiento y falta de generalización ante comportamientos no vistos. Esta tesis propone un IDS inteligente para IoT basado en arquitectura multietapa DNN¿XGBoost, diseñado para operar con datos de flujo sin inspección de contenido, optimizado para dispositivos de borde con recursos limitados. Integra: (1) DNNs especializadas por dataset (UNSW-NB15, CIC-IDS2017, etc.), (2) stacking con calibración (Platt scaling/isotonic), (3) clasificador XGBoost final. La investigación incluye: una SLR sobre 250+ estudios identificando carencias metodológicas (dependencia de un solo dataset, métricas inadecuadas, falta de validación en producción); un pipeline reproducible de preprocesado y validación con codificación, normalización y balanceo (submuestreo, SMOTE+Tomek Links); evolución desde DNN base con K-folds hasta arquitectura multietapa evaluada en múltiples datasets IoT, demostrando mejoras en estabilidad, precisión y generalización entre dominios, con robustez ante clases minoritarias e interpretabilidad sin penalizar latencia. El sistema se implementa en Jetson Nano y WSL, midiendo comportamiento extremo a extremo (captura de tráfico, extracción de flujos, inferencia casi real) y comparándolo con Suricata y Snort. Los resultados confirman viabilidad operativa, baja latencia y equilibrio detección/falsos positivos en carga variable y tráfico cifrado. Un model store versionado garantiza reproducibilidad conservando artefactos de entrenamiento, escaladores, umbrales y semillas. La tesis aporta desde la revisión crítica hasta la validación en entornos reales, resultando en un IDS ligero, preciso y adaptable que combina rigor metodológico con eficiencia práctica. Las líneas futuras incluyen adaptación ante deriva de tráfico, aprendizaje federado e integración de mecanismos explicables para auditoría.[ENG] The growth of IoT has exponentially expanded the attack surface, generating threats that traditional IDS do not address effectively due to their limitations in adaptation, high maintenance costs, and lack of generalization when facing unseen behaviors. This thesis proposes an intelligent IDS for IoT based on a multi-stage DNN¿XGBoost architecture, designed to operate with flow data without content inspection, optimized for edge devices with limited resources. It integrates: (1) specialized DNNs per dataset (UNSW-NB15, CIC-IDS2017, etc.), (2) stacking with calibration (Platt scaling/isotonic), (3) final XGBoost classifier. The research includes: an SLR on 250+ studies identifying methodological gaps (single dataset dependency, inadequate metrics, lack of production validation); a reproducible preprocessing and validation pipeline with encoding, normalization, and balancing (undersampling, SMOTE+Tomek Links); evolution from baseline DNN with K-folds to multi-stage architecture evaluated on multiple IoT datasets, demonstrating improvements in stability, accuracy, and cross-domain generalization, with robustness against minority classes and interpretability without penalizing latency. The system is implemented on Jetson Nano and WSL, measuring end-to-end behavior (traffic capture, flow extraction, near-real-time inference) and comparing it with Suricata and Snort. The results confirm operational viability, low latency, and detection/false positive balance under variable load and encrypted traffic. A versioned model store ensures reproducibility by preserving training artifacts, scalers, thresholds, and seeds. The thesis contributes from critical review to validation in real environments, resulting in a lightweight, accurate, and adaptable IDS that combines methodological rigor with practical efficiency. Future directions include adaptation to traffic drift, federated learning, and integration of explainable mechanisms for auditing.Esta tesis forma parte del proyecto TRUST Lab, Laboratorio de I+D+i en Ciberseguridad, Privacidad y Comunicaciones Seguras, financiado por la Unión Europea NextGeneration-EU, el Plan de Recuperación, Transformación y Resiliencia, a través de INCIBE.Escuela Internacional de DoctoradoUniversidad Politécnica de CartagenaTecnologías de la Información y Comunicacione
Performing expertise and emotion in AI risk discourse
Get PDFThis article examines how public messages about artificial intelligence (AI) are shaped through multimodal communication in TED Talks. Focusing on two 2023 talks on AI risks by Sasha Luccioni and Gary Marcus, the study explores how the speakers use language, visuals, and bodily performance to build credibility and engage with global audiences. Using Multimodal Discourse Analysis, we identify key “rich points”, moments where gesture, imagery, voice, and narrative work together to frame AI as both a technological and a social risk. The comparative analysis reveals that the two speakers rely on different persuasive strategies: Luccioni uses an inclusive and emotionally engaging style to encourage identification and empathy, whereas Marcus adopts a more authoritative stance that emphasises expertise and institutional responsibility. The findings demonstrate how ethos and pathos are not fixed rhetorical tools but are constructed through interaction and multimodal resources in digitally mediated spaces. By conceptualising TED Talks as spaces where technical knowledge is turned into more accessible language, this study advances research on persuasion, how expertise is built, and how multimodal communication shapes public understanding of emerging technologies
Agent-based paradigm for the self-configuration of a conceptual mechanical assembly modeling application in virtual reality
Get PDFThe immersive, multisensory experiences offered by virtual reality have been transforma tive across multiple disciplines, enhancing practical and theoretical skills while increasing user motivation and learning. On the other hand, multi-agent systems have proven to be effective in facilitating the expansion and modularity of computer systems. This paper presents an application developed in a virtual reality environment based on multi-agent systems for the conceptual design of mechanical assemblies from primitives. As a main novelty, the primitives can be defined by the user of the application from a set of models and images, and an Excel document, without the need for programming knowledge, taking advantage of the possibilities offered by multi-agent systems. In addition, for each primi tive, it is possible to define a set of geometric and dimensional modifications, as well as a set of position relations with respect to other primitives to generate mechanical assemblies
Design and development of underwater acoustic communication systems based on smart antenna technologies
Get PDF[SPA] Esta Tesis Doctoral aborda el problema del direccionamiento acústico eficiente en medios complejos, como el aire y el agua, proponiendo como solución el análisis y desarrollo de antenas acústicas de ondas de fuga (ALWA) que permiten escanear el espacio angular mediante la variación de frecuencia, sin necesidad de emplear múltiples elementos activos ni mecanismos móviles. Frente a los sistemas tradicionales, que requieren configuraciones multicanal y elevado procesamiento, las ALWAs presentan una alternativa más compacta, eficiente y de bajo consumo. A lo largo de esta tesis se diseñan y simulan diferentes configuraciones de antenas, tanto uniformes como moduladas, empleando herramientas numéricas basadas en el Método de Elementos Finitos (FEM). Asimismo, se propone un nuevo diseño funcional en entorno subacuático, compuesto por una estructura axisimétrica con guías de titanio y anillos de bismuto flexibles, que logra un acoplamiento efectivo entre la estructura y el medio. Los resultados obtenidos demuestran la capacidad de estos dispositivos para generar radiación direccional controlada, validando su potencial para aplicaciones emergentes como el posicionamiento, la comunicación submarina o el desarrollo de sistemas acústicos compactos y autónomos.[ENG] This doctoral thesis addresses the problem of efficient acoustic steering in complex media, such as air and water, proposing as a solution the analysis and development of acoustic leaky wave antennas (ALWAs) that allow angular space to be scanned by varying the frequency, without the need for multiple active elements or moving mechanisms. Compared to traditional systems, which require multi-channel configurations and high processing power, ALWAs offer a more compact, efficient, and low-power alternative. Throughout the work, different antenna configurations, both uniform and modulated, are designed and simulated using numerical tools based on the Finite Element Method (FEM). A new functional design for underwater environments is also proposed, consisting of an axisymmetric structure with titanium guides and bismuth rings, which achieves effective coupling between the structure and the medium. The results obtained demonstrate the ability of these devices to generate controlled directional radiation, validating their potential for emerging applications such as positioning, underwater communication, or the development of compact and autonomous acoustic systems.Escuela Internacional de DoctoradoUniversidad Politécnica de CartagenaTecnologías de la Información y Comunicacione
Audio fundamentals
No full textThis English-language resource introduces audio fundamentals through a multi-activity unit focused on the DIX4192-Q1 digital audio interface system and related topics. It includes reading comprehension on device architecture, specifications, and control (SPI/I2C), pronunciation of key technical terms with audio, fill-in-the-blank exercises, speaking prompts, and two watching activities on audio capacitors and amplifier basics with quizzes. A grammar section on gerunds and infinitives and a Kahoot review reinforce learning
PID controller (proportional integrative derivative)
No full textThis resource provides an English-language instructional unit on PID controllers, covering the proportional, integral, and derivative components, their mathematical formulation, tuning strategies such as Ziegler–Nichols, and applications across industry, automation, transportation, and HVAC. It includes reading comprehension, vocabulary with pronunciations and audio terms, sentence completion, class discussions, videos with quizzes, grammar practice on countable/uncountable nouns, and a debate activity contrasting open- and closed-loop control systems