e-cienciaDatos
Not a member yet
1765 research outputs found
Sort by
Balance Autoabastecimiento Legumbres Comunidad Madrid = Pulses self-supply potential in Madrid Region
No full textAssessment of the food (pulses) self-sufficiency supply capacity in the Comunidad de Madrid (Spain) with different dietary and distribution scenarios.
The dataset is detailed at the municipal level, providing the number of people in each municipality that could meet their food (pulses) needs considering a) the average legume intake and b) the dietary recommendation and with two different distribution models: a) Centralizaed in the capital city, b) descentralize
I-SUPPORT Robot: Data and results of the identification and control of a two-module soft arm
No full textThis project focuses on the I-SUPPORT soft robotic arm, a two-module continuum manipulator designed for assistive bathing tasks. A novel methodology is developed for independent position and orientation control, called Independent Distal Module Actuation (IDMA). The proximal module controls position (X,Z), while the distal module controls orientation (pitch, yaw). To cope with the nonlinear and compliant behavior of the arm, fractional-order controllers (FOPI) are implemented after performing system identification with Intel RealSense camera data. The approach is validated in both vertical and horizontal configurations, and experiments include position tracking, orientation regulation, load disturbance rejection, and task execution (e.g., sponge rubbing on the back).The dataset includes identification data collected from the Intel RealSense depth camera while actuating the modules with tendons and McKibben chambers, as well as experimental results of step responses under different load conditions. It also contains position tracking experiments in both vertical and horizontal configurations, orientation control datasets of the distal module, and task control datasets corresponding to sponge rubbing experiments. In addition, MATLAB scripts are provided for modeling, controller design, data collection, and plotting, which allow the reproduction and analysis of the experiments described
Augmented Rey-Osterrieth Complex Figure dataset (binary images) / Conjunto aumentado de datos de figuras complejas de Rey-Osterrieth (imágenes binarias)
No full text Description of the project
This dataset is the result of the research carried out in the project "Mild Cognitive Impairment (MCI) Detection from Rey-Osterrieth Complex Figure Copy (ROCF) Drawings using a Contrastive Loss Siamese Neural Network". In this project, we employed an augmented version of ROCFD528 to train a siamese neural network that was designed to distinguish ROCF drawings made by a healthy individual from those made by an individual diagnosed with MCI.
Description of the dataset
This dataset contains a single folder with 2640 binary images of Rey-Osterrieth Complex Figures.
Methodology
For each drawing in the "Rey-Osterrieth Complex Figure dataset (binary images)" (also known as ROCFD528), we generated four augmented versions using random geometric transformations (rotations up to 8 degrees, shears up to 12 degrees, and scaling with factors ranging from 0.98 to 1.15).
Descripción del proyecto
El contenido de este conjunto de datos es resultado de la investigación llevada a cabo en el proyecto "Detección de Deterioro Cognitivo Leve (DCL) partiendo de Dibujos de la Figura Compleja de Rey-Osterrieth (FCRO) usando una Red Neuronal Siamesa". En este proyecto, empleamos una versión aumentada de ROCFD528 para entrenar una red neuronal siamesa cuya función es distinguir dibujos de la FCRO efectuados por una persona sana de aquellos realizados por una persona con DCL.
Descripción del dataset
Este dataset contiene una única carpeta con 2640 imágenes binarias de figuras complejas de Rey-Osterrieth.
Metodología
Para cada dibujo del "Conjunto de datos de figuras complejas de Rey-Osterrieth (imágenes binarias)" (también llamado ROCFD528), nosotros generamos cuatro versiones aumentadas de éste empleando transformaciones geométricas aleatorias (rotaciones de hasta 8 grados, cizallados de hasta 12 grados y escalados con factores que van desde 0.98 a 1.15). </p
Functional and Structural Analysis of Respiratory Protective Filters Against Exposure to Suspended Solid Particles
No full textThe aim of this project is to perform a functional and structural analysis of respiratory protective filters exposed to suspended solid particles. The study examines how particle size and electric charge influence retention, highlighting the need to adapt filter selection to the specific dust profile of the working environment. Changes in filter porosity between classes and after use are evaluated to predict clogging processes, while reductions in air permeability following dust exposure are assessed to estimate the effective service life of the filters
Exploring CeO2-doped Co/SBA-15 catalysts for acetic acid oxi-dative steam reforming
No full textThis work explores the effect of CeO2 incorporation into Co/SBA-15 catalysts on hy-drogen production through acetic acid oxidative steam reforming as a bio-oil aqueous phase model compound. CeO2 was incorporated (5–30 wt.%) to improve the physico-chemical properties of the catalyst. XRD analysis confirmed that CeO2 addition resulted in smaller Co0 mean crystallite sizes, while H2-TPR showed enhanced reducibility properties. The catalytic performance was evaluated in the 400–700°C range, S/C molar ratio = 2, O2/C molar ratio = 0.0375, WHSV = 30.2 h-1, and P = 1 atm. Catalysts containing 10 and 20 wt.% of CeO2 exhibited the best catalytic performance, achieving nearly complete conversions and H2 yield values approaching thermodynamic equilibrium at 550 °C. Both samples maintained an acetic acid conversion above 90% after 30 h of time-on-stream with H2 yields above 55 % along the steam reforming tests. This agrees with their lower coke formation rate (7.2 and 12.0 mgcoke·gcat-1·h-1 for Co/10CeO2-SBA15 and Co/20CeO2-SBA15, respectively)
CantabricOIS2-Datos
No full textDatos del proyecto CantabricOIS2, PID2020-115192GB-I00. Inventario del registro arqueológico recuperado en las excavaciones arqueológicas realizadas durante 2022 y 2023 en la sala de la Natividad de la cueva de El Cierro (Fresnu, Ribadesella, Asturias)
Códigos TFM e Investigación Posterior - Alejandro Redondo García
No full textCódigos Deep Learning (entrenamiento y predicción) de un clasificador de la gravedad de lesiones orales - TFM Alejandro Redondo García. En los ficheros encontramos dos carpetas (si seleccionamos la visualización del árbol en la web).
1. Codigo_InvestigacionPosteriorTFM. Esta carpeta contiene lo siguiente: archivo ClasificadorLesionesOrales_CrossValidation.py - es el código en python del clasificador de Deep Learning con cross validation, archivo GenerarCSV.py - es el código en python para generar el .csv, archivo Manual.pdf - son las instrucciones para ejecutar el código ClasificadorLesionesOrales_CrossValidation.py, archivo modelosTimm.csv - son los modelos disponibles de clasificación, archivo PrediccionClasificadorLesionesOrales.py - es el código para realizar la inferencia del modelo entrenado, carpeta DatasetCross - contiene un ejemplo del formato del dataset que necesita el código ClasificadorLesionesOrales_CrossValidation.py y carpeta Resultados - es un ejemplo de resultado que arroja del modelo entrenado, concretamente se guardan las métricas, la configuración y los pesos del modelo.
2. Codigo_TFM. Esta carpeta contiene lo siguiente: archivo ClasificadorLesionesOrales.py - es el código en python del clasificador de Deep Learning sin cross validation, archivo Manual.pdf - son las instrucciones para ejecutar el código ClasificadorLesionesOrales.py, archivo modelosTimm.csv - son los modelos disponibles de clasificación, archivo PrediccionClasificadorLesionesOrales.py - es el código para realizar la inferencia del modelo entrenado, carpeta Dataset - contiene un ejemplo del formato del dataset que necesita el código ClasificadorLesionesOrales.py, carpeta Predicciones - contiene un ejemplo de la inferencia del modelo y carpeta Resultados - es un ejemplo de resultado que arroja del modelo entrenado, concretamente se guardan las métricas, la configuración y los pesos del modelo.
Nota.: Es recomendable leer los manuales antes de utilizar los códigos
Multi-objective optimization results for the splint design case study.
No full textMulti-objective optimization results for the splint design case study. Pareto Front and Analytic graphs
El alquiler de casas, cuartos y piezas en Madrid a mediados del siglo XVIII: datos georreferenciados de anuncios de prensa (1758-1768)
No full textEste dataset reúne los datos utilizados para la construcción del SIG histórico “El alquiler de casas, cuartos y piezas en Madrid a mediados del siglo XVIII”, publicado como mapaweb interactivo mediante la tecnología de ArcGIS Online. Contiene información georreferenciada de un corpus de anuncios publicados en el "Diario Noticioso, Curioso-Erudito y Comercial" entre 1758 y 1768 sobre la oferta y demanda de viviendas de alquiler en la villa y corte de Madrid, así como de alojamientos destinados a los forasteros en posadas y casas de particulares. El dataset contiene la información completa de los anuncios publicados y aporta datos sobre las tipologías de habitaciones disponibles, sus características y cualidades, localización de las viviendas, precios de arrendamiento y localización de cada casa
Corpora of Venezuelan political discourses (Maduro and Guaidó, 2019-2020)
No full textThe project's main aim is to explore the discursive construction of polarisation in two corpora of political discourses (one including speeches and interviews, and another X posts) of Nicolás Maduro and Juan Guaidó, the key figures in the Venezuelan Presidential Crisis. This event began on January 23rd, 2019, when Juan Guaidó proclaimed himself interim president of Venezuela even though Nicolás Maduro had already won the general elections in May 2018. To analyse polarisation in this context and corpus, the project looks at its evaluative collocates, polarising metaphors and social actor representations.
This sample contains two corpora: one of 5 political speeches addressed by Maduro and Guaidó and 4 political interviews targeting these leaders and carried out by the international TV channels BBC and Euronews (33547 words in total), and another of 5448 X posts (199916 words) posted in Nicolás Maduro and Juan Guaidó's official X accounts (@NicolasMaduro and @JGuaido, respectively) during the first and most active year of the presidential crisis, i.e., 2019-2020. Both original posts and retweets are included. Posts in languages other than Spanish were excluded, as well as those containing only emojis and images. As for the political speeches and interviews, the speeches were delivered by both leaders the day of Guaidó's self-proclamation (23 January 2019) and exactly a year after that (23 January 2020), and the interviews were conducted around the same dates (February and March 2019), as detailed below:
Nicolás Maduro:
Annual commemoration speech and response to Guaidó's self-proclamation (23/01/2019)
Annual commemoration speech (23/01/2020)
BBC interview (12/02/2019)
Euronews interview (13/02/2019)
Juan Guaidó:
Self-proclamation speech (23/01/2019)
World Economic Forum speech (23/01/2020)
National Conflict Resolution speech (21/02/2020)
Euronews interview (18/02/2019)
BBC interview (15/03/2019)
Each corpus was divided into two subcorpora for practical purposes, each corresponding to the politicians under study, i.e., Nicolás Maduro and Juan Guaidó.</p