Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 3
Encontrada(s) 1 página(s)

Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

An agent-based simulator applied to teaching-learning process to predict sociometric indices in higher education

  • García-Magariño, Iván
  • Plaza García, Inmaculada
  • Igual Catalán, Raúl
  • Lombas, Andrés S.
  • Jamali, Hana
Most novice teachers and even some experienced teachers can lack appropriate tools for designing teaching strategies that ensure the quality of education. The ability of working in teams is crucial in educating professionals. The literature proves that social relations influence the performance of teams. For instance, the team cohesion is directly related with its performance. In the current work, we have developed an agent-based tool for assisting teachers in simulating their teaching strategies to estimate their influence on the group sociometrics like cohesion, coherence of reciprocal relations, dissociation and density of relations. The experiments with nine scenarios in disciplines of computer science, electronic, psychology, business, tourism and renewal energies with 239 students and six teachers including experienced and novice ones show that its underlying agent-based framework can adapt to different disciplines obtaining similar outcomes to the real ones. We learned that the tool was especially reliable in predicting the density of relations and the cohesion, being the latter one probably the most relevant due to its known relation with academic performance. In addition, we also learned that it was difficult to assess the prediction quality of the dissociation in higher education, due to the usual low amounts or absence of reciprocal rejections in the students' groups in this educational stage. The presented agent-based tool is publicly distributed as open source for facilitating other researchers in following this research line.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Fog computing for assisting and tracking elder patients with neurodegenerative diseases

  • García-Magariño, I.
  • Varela-Aldas, J.
  • Palacios-Navarro, G.
  • Lloret, J.
U.S. hospitals transmit and manage great amounts of information with the avenue of Internet of things. This work departs from a real need detected by healthcare centers and hospitals in U.S., Spain and Ecuador. This work focuses on the application of fog computing for obtaining an app rich in visual content that will not overload U.S. hospital infrastructures even if it was used massively. The simulation results showed that the proposed fog-based approach could support a regular use (one day out of three on average) by 1% of patients of one of the most common neurodegenerative diseases in 14 states in U.S (i.e. 36, 400 patients in total) with only a traffic of 528 KB per day on average when using one hospital per state.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Estimation of missing prices in real-estate market agent-based simulations with machine learning and dimensionality reduction methods

  • García-Magariño, I.
  • Medrano, C.
  • Delgado, J.
The opacity of real-estate market involves some challenges in their agent-based simulation. While some real-estate Web sites provide the prices of a great amount of houses publicly, the prices of the rest are not available. The estimation of these prices is necessary for simulating their evolution from a complete initial set of houses. Additionally, this estimation could also be useful for other purposes such as appraising houses, letting buyers know which are the best offered prices (i.e., the lowest ones compared to the appraisals) and recommending the buyers to set an initial price. This work proposes combining dimensionality reduction methods with machine learning techniques to obtain the estimated prices. In particular, this work analyzes the use of nonnegative factorization, recursive feature elimination and feature selection with a variance threshold, as dimensionality reduction methods. It compares the application of linear regression, support vector regression, the k-nearest neighbors and a multilayer perceptron neural network, as machine learning techniques. This work has applied a tenfold cross-validation for comparing the estimations and errors and assessing the improvement over a basic estimator commonly used in the beginning of simulations. The developed software and the used dataset are freely available from a data research repository for the sake of reproducibility and the support to other researchers.