eBiltegia
https://ebiltegia.mondragon.edu:443/xmlui
Dspace errepositorio digitalak jaso, gorde, indexatu, preserbatu eta banatzen ditu materialak formatu digitalean.2024-03-19T13:13:54ZInfluencia de los datos biográficos y experiencia deportiva de los estudiantes universitarios de educación fisíco-deportiva sobre sus creencias, actitudes y autoconcepto físico
https://hdl.handle.net/20.500.11984/6294
Influencia de los datos biográficos y experiencia deportiva de los estudiantes universitarios de educación fisíco-deportiva sobre sus creencias, actitudes y autoconcepto físico; Influence of biographical data and sports experience of university students of physical-sports education on their beliefs, attitudes and physical self-concept
Albisua, Neritzel
El propósito de este estudio fue examinar cómo influyen las experiencias deportivas y su biografía en cada sujeto, con respecto a sus creencias y actitudes de cara a la educación física y su autoconcepto físico. Se utilizaron cuatro escalas diferentes en 826 estudiantes de 18-24 años en total que vivían en el País Vasco, encontrado por método de muestreo aleatorio simple. La primera escala es la "Escala de información personal" que consta de 4 ítems para determinar las características socio-personales, sobre la experiencia deportiva, perfl de los entrenadores/as y calificación en Educación Física. En cuanto a la segunda escala utilizada fue Feelings about Physical Education (Bourke, & Frampton, 1992) que examina las Percepciones sobre la Educación Física (PEF). La tercera escala utilizada fue el Cuestionario de actitudes hacia la educación física CAEF (Moreno, Rodriguez y Gutierrez, 2006), en su validación al Euskera de Actitudes hacia la disciplina de Educación Física (ADEF), y por último el Cuestionario AFI, Autokontzeptu Fisikoaren Itaunketa (Esnaola, 2006), que examina el autoconcepto físico de los sujetos. Los datos del estudio fueron analizados por SPSS 25 paquete de software. Se realizó un análisis de componentes principales (ACP) con el propósito de explorar la correlación entre escalas, así como también entre las escalas y las categorías de las variables biográficas y de experiencia deportiva. Se determinó que hay correlación de la gran mayoría de las escalas de los cuestionarios. En conclusión, se hayo una clara separación entre dos grupos o perfiles de estudiantes determinada principalmente por las respuestas a los cuestionarios de actitudes hacia la actividad física, creencias y autoconcepto.; The purpose of this study was to examine how sports experiences and their biography influence each subject, with respect to their beliefs and attitudes towards physical education and their physical self-concept. Four different scales were used in 826 students aged 18-24 years in total who lived in the Basque Country, found by simple random sampling method. The first scale is the "Personal Information Scale" that consists of 4 items to determine the socio-personal characteristics, the sporting experience, profile of the coaches and qualification in Physical Education. As for the second scale used, it was Feelings about Physical Education (Bourke, & Morgan, 2005) that examines Perceptions about Physical Education (PEF). The third scale used was the Questionnaire of attitudes towards physical education CAEF (Morenoet al., 2003), in its validation to the Basque of Attitudes towards the discipline of Physical Education (ADEF), and finally the AFI Questionnaire, Autokontzeptu Fisikoaren Itaunketa (Esnaola, 2006), which examines the physical self-concept of the subjects. The study data were analyzed by SPSS 25 software package. A principal component analysis (PCA) was carried out in order to explore the correlation between scales, as well as between the scales and the categories of the biographical and sports experience variables. It was determined that there is a correlation of the vast majority of the questionnaire scales. In conclusion, there is a clear separation between two groups or profiles of students determined mainly by the responses to the questionnaires on attitudes towards physical activity, beliefs and self-concept.
2023-03-06T00:00:00ZMeasurement Based Stochastic Channel Model for 60 GHz Mmwave Industrial Communications
https://hdl.handle.net/20.500.11984/6293
Measurement Based Stochastic Channel Model for 60 GHz Mmwave Industrial Communications
Osa, Joseba; Mendicute, Mikel
Communications in the mmWave spectrum are gaining relevance in the last years as they are a promising candidate to cope with the increasing demand of throughput and latency in different use cases. Nowadays, several efforts have been carried out to characterize the propagation medium of these signals with the aim of designing their corresponding communication protocols accordingly, and a wide variety of both outdoor/indoor locations have already been studied. However, very few works endorse industrial scenarios, which are particularly demanding due to their stringent requirements in terms of reliability, determinism, and latency. This work aims to provide an insight of the propagation of 60 GHz mmWave signals in a typical industrial workshop in order to explore the particularities of this kind of scenario. In order to achieve this, an extensive measurement campaign has been carried out in this environment and a stochastic channel model has been proposed and validated.
2023-01-01T00:00:00ZFederated Explainability for Network Anomaly Characterization
https://hdl.handle.net/20.500.11984/6292
Federated Explainability for Network Anomaly Characterization
Zurutuza, Urko
Machine learning (ML) based systems have shown promising results for intrusion detection due to their ability to learn complex patterns. In particular, unsupervised anomaly detection approaches offer practical advantages as does not require labeling the training data, which is costly and time-consuming. To further address practical concerns, there is a rising interest in adopting federated learning (FL) techniques as a recent ML model training paradigm for distributed settings (e.g., IoT), thereby addressing challenges such as data privacy, availability and communication cost concerns. However, output generated by unsupervised models provide limited contextual information to security analysts at SOCs, as they usually lack the means to know why a sample was classified as anomalous or cannot distinguish between different types of anomalies, difficulting the extraction of actionable information and correlation with other indicators. Moreover, ML explainability methods have received little attention in FL settings and present additional challenges due to the distributed nature and data locality requirements. This paper proposes a new methodology to characterize and explain the anomalies detected by unsupervised ML-based intrusion detection models in FL settings. We adapt and develop explainability, clustering and cluster validation algorithms to FL settings to mine patterns in the anomalous samples and identify different threats throughout the entire network, demonstrating the results on two network intrusion detection datasets containing real IoT malware, namely Gafgyt and Mirai, and various attack traces. The learned clustering results can be used to classify emerging anomalies, provide additional context that can be leveraged to gain more insight and enable the correlation of the anomalies with alerts triggered by other security solutions.
2023-01-01T00:00:00ZTowards robust defect detection in casting using contrastive learning
https://hdl.handle.net/20.500.11984/6291
Towards robust defect detection in casting using contrastive learning
Intxausti Arbaiza, Eneko; Zugasti, Ekhi; Cernuda, Carlos
Defect detection plays a vital role in ensuring product quality and safety within industrial casting processes. In these dynamic environments, the occasional emergence of new defects in the production line poses a significant challenge for supervised methods. We present a defect detection framework to effectively detect novel defect patterns without prior exposure during training. Our method is based on contrastive learning applied to the Faster R-CNN model, enhanced with a contrastive head to obtain discriminative representations of different defects. By training on an diverse and comprehensive labeled dataset, our method achieves comparable performance to the supervised baseline model, showcasing commendable defect detection capabilities. To evaluate the robustness of our approach, we authentically replicate a real-world use case by deliberately excluding several defect types from the training data. Remarkably, in this new context, our proposed method significantly improves detection performance of the baseline model, particularly in situations with very limited training data, showcasing a remarkable 34.7% enhancement. Our research highlights the potential of the proposed method in real-world environments where the number of available images may be limited or inexistent. By providing valuable insights into defect detection in challenging scenarios, our framework could contribute to ensuring efficient and reliable product quality and safety in industrial manufacturing processes.
2023-01-01T00:00:00Z