Abstract
Public opinion is nowadays a valuable data source for many sectors. In this study, we analysed the transportation sector using messages extracted from Twitter. Contrasting with the traditional surveying methods that are high-cost and inefficient used in transportation sector, social media are popular sources of crowdsensing. This work used BERT embeddings, an unsupervised pre-trained model released in 2018, to classify travel-related terms using tweets collected from three distinct cities: New York, London, and Melbourne. In order to understand if a simple model can have a good performance, we used unigrams. A list of 24 travel-related words was used to classify the messages. Popular words are train, walk, car, station, street, and avenue. Between 3% to 5% of all messages are classified as traffic-related, while along the typical working hours of the day the values is around 5-6%. A high model performance was obtained, with precision and accuracy higher than 0.80 and 0.90, respectively. The results are consistent for all the three cities assessed.

Abstract
The COVID-19 virus outbreak led to the need of developing smart disinfection systems, not only to protect the people that usually frequent public spaces but also to protect those who have to subject themselves to the contaminated areas. In this paper it is developed a human detector smart sensor for autonomous disinfection mobile robot that use Ultra Violet C type light for the disinfection task and stops the disinfection system when a human is detected around the robot in all directions. UVC light is dangerous for humans and thus the need for a human detection system that will protect them by disabling the disinfection process, as soon as a person is detected. This system uses a Raspberry Pi Camera with a Single Shot Detector (SSD) Mobilenet neural network to identify and detect persons. It also has a FLIR 3.5 Thermal camera that measures temperatures that are used to detect humans when within a certain range of temperatures. The normal human skin temperature is the reference value for the range definition. The results show that the fusion of both sensors data improves the system performance, compared to when the sensors are used individually. One of the tests performed proves that the system is able to distinguish a person in a picture from a real person by fusing the thermal camera and the visible light camera data. The detection results validate the proposed system.

Abstract

Abstract
Este estudio plantea una alternativa de consecución de los parámetros eléctricos en modelos de transformadores trifásicos mediante registros obtenidos con lecturas de aparatos de medición en sistemas de protección y monitoreo asociados al transformador eléctrico. El posible uso de este procedimiento se focaliza en la estimación del modelo eléctrico en transformadores trifásicos durante su operación y funcionamiento en un sistema eléctrico y su respuesta frente a eventos transitorios. Esta propuesta permitirá aminorar recursos y uso de tiempo en comparación con metodologías existentes; puesto que, éstas usualmente necesitan que el transformador este fuera de servicio para la conexión de equipo especializado de alto costo, siendo muchas veces necesario trasladar el transformador a un laboratorio especializado.  Los trabajos existentes proporcionan modelos completos del transformador para simulación de eventos transitorios, estos se usaron como datos de partida para sintonizar los parámetros en MATLAB-SIMULINK. Evaluada la efectividad y confiabilidad del estudio, éste será de gran ayuda en la determinación de parámetros eléctricos transitorios en transformadores nuevos o con un tiempo considerable de funcionamiento, pues debido al uso del equipo y las condiciones de operación, los parámetros eléctricos pueden variar con respecto a sus registros de fábrica.

Abstract
Predawn leaf water potential (Psi(pd)) is widely used to assess plant water status. Also, pigments concentration work as proxy of canopy's water status. Spectral data methods have been applied to monitor and assess crop's biophysical variables. This work developed two models to estimate Psi(pd) using a hand-held spectroradiometer (400-1010 nm) to obtain canopy and foliar reflectance in four dates of 2018 and a pressure chamber to measure Psi(pd). Two modelling approaches, combining spectral data and several machine learning algorithms (MLA), were used to estimate Psi(pd) in a commercial vineyard in the Douro Wine Region. The first approach estimated Psi(pd) through vine's canopy reflectance; several vegetation indices (VIs) were computed and selected, namely the SPVi(opt)(1_)(950;596;521;) SPVIopt2_896;880;901; PRI_CI2(opt_539;560,573;716 )and NPCIopt_983;972, as well as a time-dynamic variable based on Psi(pd) (Psi(pd)_(0)). The second modelling approach is based on pigments' concentrations; several VIs were optimized for non-correlated pigments of vine's leaves, assessed by its hyperspectral reflectance. The following variables for Psi(pd) estimation were selected through stepwise forward method: Psi(pd)_(0); NRIgreen_LUT520;532; NRIgreen_LWC540;551. The B-MARS algorithm performed the best results for both modelling approaches, presenting a RRMSE in both validation modelling approaches between 13-14%.

Abstract
In the past decade, we have witnessed the widespread adoption of Deep Neural Networks (DNNs) in several Machine Learning tasks. However, in many critical domains, such as healthcare, finance, or law enforcement, transparency is crucial. In particular, the lack of ability to conform with prior knowledge greatly affects the trustworthiness of predictive models. This paper contributes to the trustworthiness of DNNs by promoting monotonicity. We develop a multi-layer learning architecture that handles a subset of features in a dataset that, according to prior knowledge, have a monotonic relation with the response variable. We use two alternative approaches: (i) imposing constraints on the model's parameters, and (ii) applying an additional component to the loss function that penalises non-monotonic gradients. Our method is evaluated on classification and regression tasks using two datasets. Our model is able to conform to known monotonic relations, improving trustworthiness in decision making, while simultaneously maintaining small and controllable degradation in predictive ability.