Explore the vast range of titles available.

In palynology, the visual classification of pollen grains from different species is a hard task which is usually tackled by human operators using microscopes. Many industries, including medical and pharmaceutical, rely on the accuracy of this manual classification process, which is reported to be around 67%. In this paper, we propose a new method to automatically classify pollen grains using deep learning techniques that improve the correct classification rates in images not previously seen by the models. Our proposal manages to properly classify up to 98% of the examples from a dataset with 46 different classes of pollen grains, produced by the Classifynder classification system. This is an unprecedented result which surpasses all previous attempts both in accuracy and number and difficulty of taxa under consideration, which include types previously considered as indistinguishable.

In palynology, the visual classification of pollen grains from different species is a hard task which is usually tackled by human operators using microscopes. Its complete automatization would save a high quantity of resources and provide valuable improvements especially for allergy-related information systems, but also for other application fields as paleoclimate reconstruction, quality control of honey based products, collection of evidences in criminal investigations or fabric dating and tracking. This paper presents three state-of-the-art deep learning classification methods applied to the recently published POLEN23E image dataset. The three methods make use of convolutional neural networks: the first one is strictly based on the idea of transfer learning, the second one is based on feature extraction and the third one represents a hybrid approach, combining transfer learning and feature extraction. The results from the three methods are indeed very good, reaching over 97% correct classification rates in images not previously seen by the models, where other authors reported around 70.

High concentration episodes for NO2 are increasingly dealt with by authorities through traffic restrictions which are activated when air quality deteriorates beyond certain thresholds. Foreseeing the probability that pollutant concentrations reach those thresholds becomes thus a necessity. Probabilistic forecasting, as oposed to point-forecasting, is a family of techniques that allow for the prediction of the expected distribution function instead of a single future value. In the case of NO 2 , it allows for the calculation of future chances of exceeding thresholds and to detect pollution peaks. However, there is a lack of comparative studies for probabilistic models in the field of air pollution. In this work, we thoroughly compared 10 state of the art quantile regression models, using them to predict the distribution of NO 2 concentrations in a urban location for a set of forecasting horizons (up to 60 hours into the future). Instead of using directly the quantiles, we derived from them the parameters of a predicted distribution, rendering this method semi-parametric. Amongst the models tested, quantile gradient boosted trees show the best performance, yielding the best results for both expected point value and full distribution. However, we found the simpler quantile k -nearest neighbors combined with a linear regression provided similar results with much lower training time and complexity.

In the framework of extreme pollution concentrations being more and more frequent in many cities nowadays, air quality forecasting is crucial to protect public health through the anticipation of unpopular measures like traffic restrictions. In this work, we develop the core of a 48 h ahead forecasting system which is being deployed for the city of Madrid. To this end, we investigate the predictive power of a set of neural network models, including several families of deep networks, applied to the task of predicting nitrogen dioxide concentrations in an urban environment. Careful feature engineering on a set of related magnitudes as meteorology and traffic has proven useful, and we have coupled these neural models with mesoscale numerical pollution forecasts, which improve precision by up to 10%. The experiments show that some neural networks and ensembles consistently outperform the reference models, particularly improving the Naive model’s results from around (20%) up to (57%) for longer forecasting horizons. However, results also reveal that deeper networks are not particularly better than shallow ones in this setting.

Air pollution and airborne pollen play a key role in respiratory and circulatory disorders and thus have a direct relation to hospital admissions for these causes. Knowing in advance the influx of patients to emergency services allows clinical institutions to optimize resources and to improve their service. Since the variables influencing respiratory and circulatory-related hospital admissions belong to fields such aerobiology or meteorology, we aim for a data-based system which is able to predict admissions without a priori assumptions. Given the number and distribution of observation stations (meteorological, pollen and chemical pollution stations and hospital), previous approaches generate many model-dependent systems that need to be combined in order to obtain the full representation of future environmental conditions. A unified approach able to extract all temporal dynamics as well as all spatial relations would allow a better representation of the aforementioned conditions and consequently a more precise hospital admissions forecast. The proposed system is based on a specific neural network topology of long short-term memories and convolutional neural networks to obtain the spatio-temporal relations between all independent and target variables. It was applied to forecast daily hospital admissions due to respiratory- and circulatory-related disorders. The proposal outperforms the benchmark approaches by reducing as an average the prediction error by 28% and 20% for the circulatory and respiratory cases, respectively. Consequently, the system extracts all relevant information without specific field knowledge and provides accurate hospital admissions forecasts.

When confronting a spatio-temporal regression, it is sensible to feed the model with any available prior information about the spatial dimension. For example, it is common to define the architecture of neural networks based on spatial closeness, adjacency, or correlation. A common alternative, if spatial information is not available or is too costly to introduce it in the model, is to learn it as an extra step of the model. While the use of prior spatial knowledge, given or learned, might be beneficial, in this work we question this principle by comparing traditional forms of convolution-based neural networks for regression with their respective spatial agnostic versions. Our results show that the typical inclusion of prior spatial information is not really needed in most cases. In order to validate this counterintuitive result, we perform thorough experiments over ten different datasets related to sustainable mobility and air quality, substantiating our conclusions on real world problems with direct implications for public health and economy. By comparing the performance over these datasets between traditional and their respective agnostic models, we can confirm the statistical significance of our findings with a confidence of 95%.

Anticipating future workloads in a hospital may be of capital importance in order to distribute resources and improve patient attention. In this paper, we tackle the problem of predicting daily hospital admissions in Madrid due to circulatory and respiratory cases based on biometeorological indicators. A range of forecasting algorithms were proposed covering four model families: ensemble methods, boosting methods, artificial neural networks and ARIMA. Experiments show how the last two obtain better results in average, demonstrating that the problem can be properly solved with both approaches. Furthermore, a recently proposed technique known as stacked generalization was also used to dynamically combine the predictions from the four models, finally improving the performance with respect to the individual models.

The occurrence of earthquakes has been studied from many aspects. Apparently, earthquakes occur without warning and can devastate entire cities in just a few seconds, causing numerous casualties and huge economic loss. Great effort has been directed towards being able to predict these natural disasters, and taking precautionary measures. However, simultaneously predicting when, where and the magnitude of the next earthquake, within a limited region and time, seems an almost impossible task. Techniques from the field of data mining are providing new and important information to researchers. This article reviews the use of artificial neural networks for earthquake prediction in response to the increasing amount of recently published works and presenting claims of being effective. Based on an analysis and discussion of recent results, data mining practitioners are encouraged to apply their own techniques in this emerging field of research.

En este documento se dibuja una posición ética respecto al uso de tecnologías basadas en datos masivos, con atención particular al contexto de los procesos de enseñanza/aprendizaje en la UNED. Se aboga aquí por una toma de conciencia de los riesgos concretos que van aparejados con las ventajas esperables del uso de dichas tecnologías. Además, se esbozan algunas preguntas importantes y algunas respuestas posibles, incluyendo nueve cautelas esenciales, a fin de proporcionar unas bases sólidas a los proyectos que impliquen el uso de datos masivos en la UNED. Primero se expone lo que entendemos por una ética del cuidado, en la que la prioridad sea cuidar del alumnado a través del uso de datos y al mismo tiempo tener cuidado de que cualquier intervención basada en datos sea pulcra en sus presunciones y no introduzca sesgos. A continuación, se expone una selección de referentes y aproximaciones previas que han sido tenidas en cuenta y se enumeran las preguntas clave que deben guiar a cualquier institución que se plantee el trabajo con datos. Finalmente, se enuncian nueve cautelas esenciales que la institución se compromete a observar y que han sido propuestas a la comunidad universitaria para su deliberación.

Las tecnologías de identificación biométrica han experimentado un gran desarrollo en los últimos años, siendo aplicadas a decenas de ámbitos diferentes, entre los que se encuentra el ámbito educativo, en particular el universitario. Sin embargo, en este artículo argumentamos que dicha tendencia puede impactar de formas inesperadas a los procesos de enseñanza/aprendizaje. Así, se exponen algunas consideraciones principales acerca del uso de tecnologías de identificación biométrica en general y más particularmente de técnicas de reconocimiento facial en el marco de los exámenes universitarios, prestando especial atención al problema de realizar los exámenes presenciales por medios remotos durante la pandemia de la COVID-19. Se ofrece un análisis general de las limitaciones de esta tecnología en sus dimensiones técnica, jurídica y ética, y se exploran las posibles consecuencias, en esas tres dimensiones, del uso de dichas tecnologías. A modo de ilustración, se toma el caso concreto de la UNED, la universidad más grande del Estado español, exponiendo las decisiones tomadas por esta institución para dar respuesta al desafío que supusieron las restricciones de movimiento durante el estado de alarma. Dado el gran número de evidencias de que esta tecnología está aquejada de graves problemas con impredecibles consecuencias, en todo caso se recomienda observar el principio de precaución y tomar decisiones con la máxima cautela.