Omar Ramirez-Perez

The classification of activities of daily living in the home of older adults makes it possible to identify risk situations and changes in behavior that may be associated with some type of problem. This information allows caregivers and health professionals to take action when these types of situations are detected. Although many machine learning classification techniques have been proposed, the effectiveness of the solution in a real-world context remains unclear in most cases due to the large number of sensors required, the type of sensors used which may pose privacy issues, and the assumption of considering only segmented sensor events for each activity before training the models. This paper presents an evaluation of different machine learning techniques using fixed time windows to extract spatio-temporal features and classify 10 human activities in a real smart home with unobtrusive sensors using the Aruba CASAS dataset. The three classification techniques that achieved better performance were Random Forest, XGBoost and Support Vector Machine, achieving an accuracy of 97% with our best model, outperforming other approaches from the literature that were using the same dataset under similar conditions. The proposed classification techniques were also evaluated under a more realistic scenario by reducing the amount of hardware required and using an additional class labeled "Other" to consider all raw sensor events, including those that do not belong to any specific activity, achieving an accuracy of 89%, outperforming other approaches from the literature using the same dataset under similar conditions.

Desde inicios de este siglo, una corriente de los videojuegos serios, denominada en este texto como videojuegos críticos, se ha concentrado en usar estos artefactos como una herramienta para comprender problemas sociales; sin embargo, para aprender sobre la capacidad del videojuego para comunicar ideas sobre nuestra realidad social es necesario saber cómo diseñarlos. Aunque el diseño participativo es una práctica creciente en el diseño de dispositivos tecnológicos, experiencias previas han señalado algunas dificultades de esta metodología para el diseño de videojuegos serios. Este artículo argumenta que, para el caso de los videojuegos críticos, esta práctica permite ilustrar relaciones de poder y darles voz a actores desplazados del discurso de los medios de comunicación tradicionales. Se presenta el desarrollo de un videojuego sobre acoso sexual en transporte público como estudio de caso, desarrollado por un equipo de profesionales expertos en conjunto con colectivos feministas y víctimas de acoso. Bajo una perspectiva de investigación-creación, se analiza el proceso de diseño seguido y los hallazgos que emergieron durante su desarrollo y después de su lanzamiento, se ilustran los conceptos de videojuego crítico como un tipo de videojuego serio, se describe el problema de acoso sexual en el marco se discusiones en el ciberfeminismo, y se explica el desarrollo metodológico del diseño del videojuego señalado. Se sugieren el grupo focal y talleres creativos como herramientas de diseño y se señalan algunos de los aspectos más importantes para el desarrollo de un videojuego serio desde esta perspectiva.

This article presents the development of an algorithm capable of detecting and classifying electric faults in photovoltaic array systems by measuring the voltage-power curve. The algorithm was build based on a characterization method in which multiple photovoltaic arrays were evaluated under different fault conditions, by measuring and analyzing the voltage-power curves at the output of each array. The algorithm was evaluated experimentally in a controlled environment inside a laboratory under 59 different fault conditions obtaining an effectiveness of 100%. Then, the algorithm was evaluated experimentally outdoors under 124 different fault conditions, temperature and solar radiations, and was able to detect and classify electric faults in different photovoltaic arrays with an effectiveness of 94.4%. The proposed algorithm can be implemented with standard power-inverters as a low-cost solution and users can receive information on up-to date performance of their photovoltaic array systems through a mobile App. The design of a mobile app for the algorithm is proposed as well.

Meditation is a form of mental training that has therapeutic potential and cognitive benefits that may enhance attention, mental well-being, and neuroplasticity. However, the learning process is not easy because meditators do not receive immediate feedback that lets them know if they are correctly doing the activity. EEG Neurofeedback training is one of the techniques to train brain self-regulation and it has the potential to increase the effectiveness of meditation. However, the benefits greatly differ between subjects with a high percentage of inefficacy. In this work, an EEG Neurofeedback Training System is proposed based on user-centered design methodology to provide real-time performance feedback to meditators to increase levels of attention and relaxation through a visual, sound and smell stimuli interface. Levels of attention and relaxation of nine participants were measured with a mobile Neurosky EEG headset biosensor during meditation practice to analyze the incidence of each type of stimuli during activity. Visual stimuli feedback was able to increase attention levels of 78% of the participants by 11.8% compared to a meditation session without any stimuli. The sound stimuli feedback was able to increase the relaxation levels of 44.4% of the participants by 16% compared to a session without any stimuli. These results might bring new insights for the design of a neurofeedback system interface for meditation. Further research on neurofeedback training interfaces for meditators is suggested to validate these results with more participants.