Search

Energy management systems allow the Smart Grids industry to track, improve, and regulate energy use. Particularly, demand-side management is regarded as a crucial component of the entire Smart Grids system. Therefore, by aligning utility offers with customer demand, anticipating future energy demands is essential for regulating consumption. An updated examination of several forecasting techniques for projecting energy short-term load forecasts is provided in this article. Each class of algorithms, including statistical techniques, Machine Learning, Deep Learning, and hybrid combinations, are comparatively evaluated and critically analyzed, based on three real consumption datasets from Spain, Germany, and the United States of America. To increase the size of tiny training datasets, this paper also proposes a data augmentation technique based on Generative Adversarial Networks. The results show that the Deep Learning-hybrid model is more accurate than traditional statistical methods and basic Machine Learning procedures. In the same direction, it is demonstrated that more comprehensive datasets assisted by complementary data, such as energy generation and weather, may significantly boost the accuracy of the models. Additionally, it is also demonstrated that Generative Adversarial Networks-based data augmentation may greatly improve algorithm accuracy. [ABSTRACT FROM AUTHOR]

This paper investigates the innovative intensity (iMID effect) in the local production systems (LPSs) using dynamic territorial units that modifies their geographical boundaries and specialization over time. The paper is based on the idea that the local communities are the real unit of generation of innovation and that they are constantly changing, behaving like a 'living innovation machine'. To explore this dynamic relationship, the paper applies the Marshallian industrial district (MID) as an approach to industrial change. Then, it focuses on the Marshallian industrial districts (MIDs) as a type of LPS and asks how the innovation effect changes in MIDs and other types of LPSs when dynamic territorial units are used. The differences in innovation intensity between the different types of LPSs are explained for Spain in the period 1991–2014 using new methods of analysis that combine causal analysis, a variable and adaptive geometry of territorial units and industrial specialization, and machine learning methods. The results show that the transformation of an LPS into another type of LPS does not automatically imply a change in innovative capacity, and the type of LPS of origin continues to be relevant in explaining current innovation processes. [ABSTRACT FROM AUTHOR]

Respiratory viruses, such as COVID-19, are spread over time and space based on human-to-human interactions. Human mobility plays a key role in the propagation of the virus. Different types of sensors in smart cities are able to continuously monitor traffic-related human mobility, showing the impact of COVID-19 on traffic volumes and patterns. In a similar way, traffic volumes measured by smart traffic sensors provide a proxy variable to capture human mobility, which is expected to have an impact on new COVID-19 infections. Adding traffic data from smart city sensors to machine learning models designed to estimate upcoming COVID-19 incidence values should provide optimized results compared to models based on COVID-19 data alone. This paper proposes a novel model to extract spatio-temporal patterns in the spread of the COVID-19 virus for short-term predictions by organizing COVID-19 incidence and traffic data as interrelated temporal sequences of spatial images. The model is trained and validated with real data from the city of Madrid in Spain for 84 weeks, combining information from 4372 traffic measuring points and 143 COVID-19 PCR test centers. The results are compared with a baseline model designed for the extraction of spatio-temporal patterns from COVID-19-only sequences of images, showing that using traffic information enhances the results when forecasting a new wave of infections (MSE values are reduced by a 70% factor). The information that traffic data has on the spread of the COVID-19 virus is also analyzed, showing that traffic data alone is not sufficient for accurate COVID-19 forecasting. [ABSTRACT FROM AUTHOR]

Spain is the third-largest producer of pork meat in the world, and many farms in several regions depend on the evolution of this market. However, the current pricing system is unfair, as some actors have better market information than others. In this context, historical pricing is an easy-to-find and affordable data source that can help all agents to be better informed. However, the time lag in data acquisition can affect their pricing decisions. In this paper, we study the effect that data acquisition delay has on a price prediction system using multiple prediction algorithms. We describe the integration of the best proposal into a decision support system prototype and test it in a real-case scenario. Specifically, we use public data from the most important regional pork meat markets in Spain published by the Ministry of Agriculture with a two-week delay and subscription-based data of the same markets obtained on the same day. The results show that the error difference between the best public and data subscription models is 0.6 Euro cents in favour of the data without delay. The market dimension makes these differences significant in the supply chain, giving pricing agents a better tool to negotiate market prices. [ABSTRACT FROM AUTHOR]

The alternation between cloudy and clear skies alters the photovoltaic production. This makes it necessary to anticipate these disturbances hours in advance for the correct operation of the electricity distribution plants and networks. In this paper, two short-term forecasting models (3 h) are developed to forecast the photovoltaic production in an integrated plant in the CIESOL building of the University of Almería. The methodology used is based on sky camera images and Artificial Intelligence techniques. Two models have been developed and compared applying artificial neural network (ANN) and support vector machine (SVM) techniques. The global irradiance predicted using sky camera images is used as an input variable in both models. In addition, the operational status of the plants has been included as an input parameter through the performance ratio. The results have shown that the errors made by ANN and SVM are very similar. For all sky conditions, the uncertainty of the production forecast differs by less than 2% from the uncertainty of the solar resource, which is the main source of error in the production models developed. [ABSTRACT FROM AUTHOR]

This paper presents the implementation of an automatic method for the reconstruction of 3D building maps. The core innovation of the proposed method is the supplementation of OpenStreetMap data with LiDAR data to reconstruct 3D urban environments automatically. The only input of the method is the area that needs to be reconstructed, defined by the enclosing points in terms of the latitude and longitude. First, area data are requested in OpenStreetMap format. However, there are certain buildings and geometries that are not fully received in OpenStreetMap files, such as information on roof types or the heights of buildings. To complete the information that is missing in the OpenStreetMap data, LiDAR data are read directly and analyzed using a convolutional neural network. The proposed approach shows that a model can be obtained with only a few samples of roof images from an urban area in Spain, and is capable of inferring roofs in other urban areas of Spain as well as other countries that were not used to train the model. The results allow us to identify a mean of 75.57% for height data and a mean of 38.81% for roof data. The finally inferred data are added to the 3D urban model, resulting in detailed and accurate 3D building maps. This work shows that the neural network is able to detect buildings that are not present in OpenStreetMap for which in LiDAR data are available. In future work, it would be interesting to compare the results of the proposed method with other approaches for generating 3D models from OSM and LiDAR data, such as point cloud segmentation or voxel-based approaches. Another area for future research could be the use of data augmentation techniques to increase the size and robustness of the training dataset. [ABSTRACT FROM AUTHOR]

This paper presents a new approach based on the combination of machine learning techniques, in particular, sentiment analysis using lexicons, and multivariate statistical methods to assess the evolution of social mood through the COVID-19 vaccination process in Spain. Analysing 41,669 Spanish tweets posted between 27 February 2020 and 31 December 2021, different sentiments were assessed using a list of Spanish words and their associations with eight basic emotions (anger, fear, anticipation, trust, surprise, sadness, joy and disgust) and three valences (neutral, negative and positive). How the different subjective emotions were distributed across the tweets was determined using several descriptive statistics; a trajectory plot representing the emotional valence vs narrative time was also included. The results achieved are highly illustrative of the social mood of citizens, registering the different emerging opinion clusters, gauging public states of mind via the collective valence, and detecting the prevalence of different emotions in the successive phases of the vaccination process. The present combination in formal models of objective and subjective information would therefore provide a more accurate vision of social reality, in this case regarding the COVID-19 vaccination process in Spain, which will enable a more effective resolution of problems. [ABSTRACT FROM AUTHOR]

The increased penetration of renewable energy sources (RESs) as an effective mechanism to reduce carbon emissions leads to an increased weather dependency for power and energy systems. This has created dynamic operation and degradation phenomena, which affect the lifetime estimation of the assets operated with RESs. For the reliable and efficient operation of RES it is crucial to monitor the health of its constituent components and feature selection is a crucial step for building robust and accurate health monitoring approaches. In this context, this paper presents a probabilistic feature selection approach, which probabilistically weights and selects features through a heuristic and iterative process for an improved asset lifetime estimation. Power transformers are key power grid assets and they are used to demonstrate the validity and impact of the proposed approach. The approach is tested on two different photovoltaic power plants operated in Spain and Australia. Results consistently show that the proposed feature-selection approach reduces the prediction error and consistently selects relevant features. The approach has been applied to transformer lifetime estimation, but it can be generally applied to assist in the lifetime estimation of other components operated in RESs. Part of the studies presented here as well as source codes are all open-source under the GitHub repository https://github.com/iramirezg/FeatureSelection. • Probabilistic feature selection approach for improved asset lifetime estimation. • Integration of environmental features for improved renewable-operated asset lifetime. • Systematic and robust feature weighting methodology. • Improved transformer lifetime estimation including sensor and environmental data. • Validated on two real photovoltaic power plant case studies. [ABSTRACT FROM AUTHOR]

In wastewater treatment plants (WWTPs), the stochastic nature of influent wastewater and operational and weather conditions cause fluctuations in effluent quality. Data-driven models can forecast effluent quality a few hours ahead as a response to the influent characteristics, providing enough time to adjust system operations and avoid undesired consequences. However, existing data for training models are often incomplete and contain missing values. On the other hand, collecting additional data by installing new sensors is costly. The trade-off between using existing incomplete data and collecting costly new data results in three data challenges faced when developing data-driven WWTP effluent forecasters. These challenges are to determine important variables to be measured, the minimum number of required data instances, and the maximum percentage of tolerable missing values that do not impede the development of an accurate model. As these issues are not discussed in previous studies, in this research, for the first time, a comprehensive analysis is done to provide answers to these challenges. Another issue that arises in all data-driven modeling is how to select an appropriate forecasting model. This paper addresses these issues by first testing nine machine learning models on data collected from three wastewater treatment plants located in Iran, Australia, and Spain. The most accurate forecaster, Bayesian network, was then used to address the articulated challenges. Key variables in forecasting effluent characteristics were flow rate, total suspended solids, electrical conductivity, phosphorus compounds, wastewater temperature, and air temperature. A minimum of 250 samples was needed during the model training to achieve a great reduction in the forecasting error. Moreover, a steep increase in the error was observed should the portion of missing values exceed 10%. The results assist plant managers in estimating the necessary data collection effort to obtain an accurate forecaster, contributing to the quality of the effluent. [Display omitted] • Bayesian network is benchmarked against eight AI models to predict wastewater quality. • Bayesian networks can handle missing value in input data due to probabilistic nature. • Key variables in WWTP quality and quantity were specified. • Input data quality has a huge role in successful development of data-driven models. • The prediction error sharply increases when the portion of missing data exceeds 10%. [ABSTRACT FROM AUTHOR]

Extreme Wind Speed events (EWS) are responsible for the worst damages caused by wind in wind farms. An accurate estimation of the frequency and intensity of EWS is essential to avoid wind turbine damage and to minimize cut-out events in these facilities. In this paper we discuss how generative Data Augmentation (DA) techniques improve the performance of Machine Learning (ML) and Deep Learning (DL) algorithms in EWS prediction problems. These problems are usually tackled as classification tasks, which are highly unbalanced due to the small number of EWS events in wind farms. Different versions of Variational AutoEncoders (VAE) are proposed and analysed in this work (VAEs, Conditional VAEs (CVAEs) and Class-Informed VAEs (CI-VAE)) as generative DA techniques to balance the data in EWS problems, leading to better performance of the prediction systems. The proposed generative DA techniques have been compared against traditional DA algorithms in a real problem of EWS prediction in Spain, considering ERA5 reanalysis data as predictive variables. The results showed that the CI-VAE with a Convolutional Neural Network approach obtained the best results, with values of Precision 0.62, Recall 0.74 and F1 score 0.67, improving up to 4% the results of the method without data augmentation techniques. [Display omitted] • Generative data augmentation techniques can improve extreme wind prediction. • Variational Autoencoders modifications are used to generate synthetic wind extreme samples. • Experiments on real wind speed data from a Spanish wind farm are presented. • Comparisons with traditional data augmentation techniques show prediction improvements. [ABSTRACT FROM AUTHOR]

[Display omitted] • Inclusion of temporal data can improve detection F 1 -Score by 8% • Infrared bands provide about 4% more F 1 -Score than visible bands despite lower resolution • Using over 80 features provides an increase of about 12% F 1 -Score over using less than 10. • The proposed method successfully detects all test plots with nearly 90% F 1 -Score • A dataset of recent manure application, verified through on-site validation, is made public. The use of manure in agricultural fields during the wet season can lead to environmental pollution by releasing nitrates into nearby water sources. To address this issue, authorities may impose closed periods during which manure application is prohibited. However, ensuring compliance with these regulations can be challenging, as it is difficult to monitor all fields in a country. To tackle this problem, a solution has been proposed that involves employing machine learning techniques in conjunction with satellite imagery to automatically identify freshly manured fields. This paper investigates the relationship and effectiveness of the Sentinel-2 satellite bands and 51 frequently utilized multispectral indices in the context of precision agriculture, by exploring different feature selection methods. The proposed method achieves nearly 90% F 1 -Score and detects all test plots of the northern Spanish region, showing its potential for large-scale use in precision agriculture and environmental monitoring. This method incorporates temporal data, resulting in an 8% improvement in the detection F 1 -Score. Despite their lower spatial resolution, infrared bands have proven to be more effective than visible bands, enhancing the F 1 -Score by 4%. Furthermore, the use of over 80 features contributes to a 12% increase in the F 1 -Score compared to using fewer than 10 features. For further research and future studies, a dataset of recently manured plots, verified on-site, has been developed and made publicly available. [ABSTRACT FROM AUTHOR]

This paper investigates the possibility of the implementation of Genetic Programming (GP) algorithm on a publicly available COVID-19 data set, in order to obtain mathematical models which could be used for estimation of confirmed, deceased, and recovered cases and the estimation of epidemiology curve for specific countries, with a high number of cases, such as China, Italy, Spain, and USA and as well as on the global scale. The conducted investigation shows that the best mathematical models produced for estimating confirmed and deceased cases achieved R2 scores of 0.999, while the models developed for estimation of recovered cases achieved the R2 score of 0.998. The equations generated for confirmed, deceased, and recovered cases were combined in order to estimate the epidemiology curve of specific countries and on the global scale. The estimated epidemiology curve for each country obtained from these equations is almost identical to the real data contained within the data set [ABSTRACT FROM AUTHOR]

This paper analyses the influence of the variation of meteorological and operational parameters on estimation of the power output of a dispatchable wind farm (WF). The active power set-points (APSPs), established to regulate the wind farm power output (WFPO), are one of the analysed parameters. The WF considered as case study is integrated in the Gorona del Viento wind-hydro power plant (El Hierro-Canary Islands-Spain), which supplies the primary energy demand of the island. Statistical inference between the errors generated by different WFPO estimation models, each fed with different input features, is performed. These models are based on supervised machine learning (ML) regression algorithms, namely support vector regression, random forest, and a combination of the strengths of these two base learning algorithms constructed using stacked regression ensemble techniques. From the results obtained, the following conclusions are drawn: a) There is a marked difference between the errors obtained with the model that only considers wind speed and direction and that which additionally incorporates the APSP parameter, showing the importance of considering this particular parameter; b) the model which incorporates air density and turbulence intensity in addition to the APSP improves the values of all the metrics, independently of the ML technique employed. • The active power set-point (APSP) parameter is very important in WFPO estimation. • Air density, turbulence intensity and APSP parameters improve WFPO estimation. • Eight models are compared for wind farm power output (WFPO) estimation. • Three machine learning techniques implemented in the models are analysed. • Non-parametric statistical inference is used to compare the models. [ABSTRACT FROM AUTHOR]

The traditional pottery industry was an important activity in Catalonia (NE Spain) up to the 20th century. However, nowadays only few workshops persist in small villages were the activity is promoted as a touristic attraction. The preservation and promotion of traditional pottery in Catalonia is part of an ongoing strategy of tourism diversification that is revitalizing the sector. The production of authenticable local pottery handicrafts aims at attracting cultivated and high-purchasing power tourists. The present paper inspects several approaches to set up a scientific protocol based on the chemical composition of both raw materials and pottery. These could be used to develop a seal of quality and provenance to regulate the sector. Six Catalan villages with a renowned tradition of local pottery production have been selected. The chemical composition of their clays and the corresponding fired products has been obtained by Energy dispersive X-ray fluorescence (EDXRF). Using the obtained geochemical dataset, a number of unsupervised and supervised machine learning methods have been applied to test their applicability to define geochemical fingerprints that could allow inter-site discrimination. The unsupervised approach fails to distinguish samples from different provenances. These methods are only roughly able to divide the different provenances in two large groups defined by their different SiO2 and CaCO3 concentrations. In contrast, almost all the tested supervised methods allow inter-site discrimination with accuracy levels above 80%, and accuracies above 85% were obtained using a meta-model combining all the predictive supervised methods. The obtained results can be taken as encouraging and demonstrative of the potential of the supervised approach as a way to define geochemical fingerprints to track or attest the provenance of samples. [ABSTRACT FROM AUTHOR]

Through the application of machine learning techniques, this paper aims to estimate the importance of messages with ideological load during the elections held in Spain on May 24th, 2015 posted by Twitter's users, as well as other variables associated with the publication of these types of messages. Our study collected and analysed 24,900 tweets associated to two of the main trending topics' hashtags (#24M and #Elections2015) used in the election day and build a predictive model to infer the ideological orientation for the messages which made use of these hashtags during Election Day. This approach allows us to classify the ideological orientation of all collected tweets, instead of only tweets that explicitly express their ideological or partisan preferences in the messages. Using the ideological orientation for all tweets predicted by our model, it was possible to identify how messages with a defined ideological load were pushed forward by users with leftist tendencies. We also observed a relationship between these messages and the partisan orientation of those who published them. [ABSTRACT FROM AUTHOR]

In this paper we propose and discuss different Deep Learning-based ensemble algorithms for a problem of low-visibility events prediction due to fog. Specifically, seven different Deep Learning (DL) architectures have been considered, from which multiple individual learners are generated. Hyperparameters of the models, including parameters concerning data preprocessing, models architecture and training procedure, are randomly selected for each model within a pre-defined discrete range. Also, every model is trained with slightly different data sampled randomly, assuring that every models introduce variety in the ensemble. Then, three different information fusion techniques are employed to build the ensemble models. The influence of the filtering process and the elitism level (the percentage of the individual models entering the ensemble) is also assessed. The performance of the proposed methodology have been tested in two real problems of low-visibility events prediction due to orographical and radiation fog, at the north of Spain. Comparison with different Machine Learning, alternative DL algorithms and meteorological-based methods show the good performance of the proposed deep learning ensembles in this problem. [ABSTRACT FROM AUTHOR]

The spiraling growth of the world's population and unregulated urbanization have resulted in many environmental problems, including poor quality of air, which is associated with a wide range of health issues. Machine learning approaches have been extensively employed to predict air quality, attracting the attention of the scientific community worldwide. Bibliometric studies provide a useful means by which to visualize and analyze published works, helping researchers to make novel scientific contributions by filling existing knowledge gaps in the research. To acquire an in-depth understanding of the topic, this paper presents a bibliometric analysis of all published articles on the use of machine learning networks to predict air quality found in the Web of Science (WoS) search engine from 1992 to 2021. S-curve analysis and social network analysis were used to identify the temporal distribution of articles, productivity by countries/continents, research institutions, and scientific metrics of journal productivity. This study indicated that maximum expansion of the literature witnessed during 2017–2021 (second phase) which represents an expansion or growth stage of machine learning and air quality prediction research. The number of published works increased significantly with 1432 articles accounting for 68.51% of all publications. As a result of the increased interest in machine learning-based prediction tools, the number of articles grew 2.17-fold compared to the 1992–2016 (first phase). In terms of international collaboration impact, Italy emerged as the most successful country (43.44), followed by Greece (31.22) and Spain (23.29). Author keywords analysis was employed to explore and evaluate the emerging research trends on the subject of air quality using machine learning models. Keywords that appear most frequently in this study are 'air pollution', 'air quality', 'machine learning', and 'forecasting'. Citation burst analysis, research productivity analysis, highly influential and highly cited works were also employed to examine various research themes and questions. In this study we also discussed how conventional methods were transformed into machine learning approaches. It is expected that this paper will provide technical guidelines, research priorities, and future opportunities for the precise prediction of air quality and emergency management of air pollution globally. • Interest in air quality prediction through machine learning model is expanding. • China and US were found productive countries in terms of countries collaboration. • Big data, Internet of Things, and Smart city are emerging topic for future research. • Significant difference between citations, funded and non-funded research. • Ensemble modeling is popular in air quality prediction research. [ABSTRACT FROM AUTHOR]

The spiraling growth of the world's population and unregulated urbanization have resulted in many environmental problems, including poor quality of air, which is associated with a wide range of health issues. Machine learning approaches have been extensively employed to predict air quality, attracting the attention of the scientific community worldwide. Bibliometric studies provide a useful means by which to visualize and analyze published works, helping researchers to make novel scientific contributions by filling existing knowledge gaps in the research. To acquire an in-depth understanding of the topic, this paper presents a bibliometric analysis of all published articles on the use of machine learning networks to predict air quality found in the Web of Science (WoS) search engine from 1992 to 2021. S-curve analysis and social network analysis were used to identify the temporal distribution of articles, productivity by countries/continents, research institutions, and scientific metrics of journal productivity. This study indicated that maximum expansion of the literature witnessed during 2017–2021 (second phase) which represents an expansion or growth stage of machine learning and air quality prediction research. The number of published works increased significantly with 1432 articles accounting for 68.51% of all publications. As a result of the increased interest in machine learning-based prediction tools, the number of articles grew 2.17-fold compared to the 1992–2016 (first phase). In terms of international collaboration impact, Italy emerged as the most successful country (43.44), followed by Greece (31.22) and Spain (23.29). Author keywords analysis was employed to explore and evaluate the emerging research trends on the subject of air quality using machine learning models. Keywords that appear most frequently in this study are 'air pollution', 'air quality', 'machine learning', and 'forecasting'. Citation burst analysis, research productivity analysis, highly influential and highly cited works were also employed to examine various research themes and questions. In this study we also discussed how conventional methods were transformed into machine learning approaches. It is expected that this paper will provide technical guidelines, research priorities, and future opportunities for the precise prediction of air quality and emergency management of air pollution globally. • Interest in air quality prediction through machine learning model is expanding. • China and US were found productive countries in terms of countries collaboration. • Big data, Internet of Things, and Smart city are emerging topic for future research. • Significant difference between citations, funded and non-funded research. • Ensemble modeling is popular in air quality prediction research. [ABSTRACT FROM AUTHOR]

Given the forthcoming availability of Sentinel-2 (S2) images, this paper provides a systematic comparison of retrieval accuracy and processing speed of a multitude of parametric, non-parametric and physically-based retrieval methods using simulated S2 data. An experimental field dataset (SPARC), collected at the agricultural site of Barrax (Spain), was used to evaluate different retrieval methods on their ability to estimate leaf area index (LAI). With regard to parametric methods, all possible band combinations for several two-band and three-band index formulations and a linear regression fitting function have been evaluated. From a set of over ten thousand indices evaluated, the best performing one was an optimized three-band combination according to ( ρ 560 - ρ 1610 - ρ 2190 ) / ( ρ 560 + ρ 1610 + ρ 2190 ) with a 10-fold cross-validation R CV 2 of 0.82 ( RMSE CV : 0.62). This family of methods excel for their fast processing speed, e.g., 0.05 s to calibrate and validate the regression function, and 3.8 s to map a simulated S2 image. With regard to non-parametric methods, 11 machine learning regression algorithms (MLRAs) have been evaluated. This methodological family has the advantage of making use of the full optical spectrum as well as flexible, nonlinear fitting. Particularly kernel-based MLRAs lead to excellent results, with variational heteroscedastic (VH) Gaussian Processes regression (GPR) as the best performing method, with a R CV 2 of 0.90 ( RMSE CV : 0.44). Additionally, the model is trained and validated relatively fast (1.70 s) and the processed image (taking 73.88 s) includes associated uncertainty estimates. More challenging is the inversion of a PROSAIL based radiative transfer model (RTM). After the generation of a look-up table (LUT), a multitude of cost functions and regularization options were evaluated. The best performing cost function is Pearson’s χ -square. It led to a R 2 of 0.74 ( RMSE : 0.80) against the validation dataset. While its validation went fast (0.33 s), due to a per-pixel LUT solving using a cost function, image processing took considerably more time (01:01:47). Summarizing, when it comes to accurate and sufficiently fast processing of imagery to generate vegetation attributes, this paper concludes that the family of kernel-based MLRAs (e.g. GPR) is the most promising processing approach. [ABSTRACT FROM AUTHOR]

Energy demand prediction is an important problem whose solution is evaluated by policy makers in order to take key decisions affecting the economy of a country. A number of previous approaches to improve the quality of this estimation have been proposed in the last decade, the majority of them applying different machine learning techniques. In this paper, the performance of a robust hybrid approach, composed of a Variable Neighborhood Search algorithm and a new class of neural network called Extreme Learning Machine, is discussed. The Variable Neighborhood Search algorithm is focused on obtaining the most relevant features among the set of initial ones, by including an exponential prediction model. While previous approaches consider that the number of macroeconomic variables used for prediction is a parameter of the algorithm (i.e., it is fixed a priori), the proposed Variable Neighborhood Search method optimizes both: the number of variables and the best ones. After this first step of feature selection, an Extreme Learning Machine network is applied to obtain the final energy demand prediction. Experiments in a real case of energy demand estimation in Spain show the excellent performance of the proposed approach. In particular, the whole method obtains an estimation of the energy demand with an error lower than 2%, even when considering the crisis years, which are a real challenge. [ABSTRACT FROM AUTHOR]

One of the main lines of research in the discipline of mobility mining is the development of predictors able to anticipate human travel behaviour in great detail. However, access to the high-resolution spatio-temporal data on which most existing solutions are based is rather limited due to multiple factors, e.g. costly access to third-party data. These restrictions give rise to a problem of developing predictors of human mobility in most setting, since the amount of data available to train these prediction models is insufficient. This paper explores the feasibility of using a public data source such as Twitter to predict the number of trips at the nationwide level. The proposed approach combines a large set of geotagged Twitter posts with an open data source published by the Spanish government on traveller mobility based on mobile phone location. Both datasets are used as input to Machine Learning models to validate the use of Twitter data for improving the prediction of these models. The results show that Twitter data have considerable value as a predictor of large-scale human mobility, especially for Long Short-Term Memory (LSTM) models. As a result, the relevance of this work resides in demonstrating that the use of Twitter could be considered as an alternative to substantially enhance the prediction of mobility within a country when it is combined with other open data sources. • A novel approach for nation-wide human mobility forecasting is proposed. • The predictor combines mobile-phone and Twitter location data as input variables. • LSTM and VAR models have been tested as candidates to develop the mechanism. • Twitter has proved to be a suitable exogenous input. • The system has been tested in Spain achieving a 6.85 CRMSE for 7-day prediction. [Display omitted] [ABSTRACT FROM AUTHOR]

Abstract: In recent years, fuzzy models have been acknowledged as a suitable approach for species distribution modelling due to their transparency and their ability to incorporate the ecological gradient theory. Specifically, the overlapping class boundaries of a fuzzy model are similar to the transitions between different environmental conditions. However, the need for ecological expert knowledge is an important constraint when applying fuzzy species distribution models. Moreover, the consistency of the ecological preferences of some fish species across different rivers has been widely contested. Recent research has shown that data-driven fuzzy models may solve this ‘knowledge acquisition bottleneck’ and this paper is a further contribution. The aim was to analyse the brown trout (Salmo trutta fario L.) habitat preferences based on a data-driven fuzzy modelling technique and to compare the resulting fuzzy models with a commonly applied modelling technique, Random Forests. A heuristic nearest ascent hill-climbing algorithm for fuzzy rule optimisation and Random Forests were applied to analyse the ecological preferences of brown trout in 93 mesohabitats. No significant differences in model performance were observed between the optimal fuzzy model and the Random Forests model and both approaches selected river width, the cover index and flow velocity as the most important variables describing brown trout habitat suitability. Further, the fuzzy model combined ecological relevance with reasonable interpretability, whereas the transparency of the Random Forests model was limited. This paper shows that fuzzy models may be a valid approach for species distribution modelling and that their performance is comparable to that of state-of-the-art modelling techniques like Random Forests. Fuzzy models could therefore be a valuable decision support tool for river managers and enhance communication between stakeholders. [Copyright &y& Elsevier]

This paper studies the economically significant problem of the optimization of customs fraud control, which is a critical issue for many countries. The European Union (EU) alone handles 4693 tons of goods every minute (2018 figures). Even though 70% of goods are imported at zero tariff, the EU raised EUR 25.4 billions in 2018, and customs-related income transferred by member states to the EU accounts for nearly 13% of its overall budget. In this field, (a) the conflicting objectives are qualitative and cannot be reduced to a common measure (security and terrorism, health, drug market access control, taxes, etc.); (b) each submitted item has dozens of characteristics; (c) there are constraints; and (d) risk analysis systems have to make decisions in real time. Although the World Customs Organization has promoted the use of artificial intelligence to increase the precision of controls, the problem is very complex due to the data characteristics and interpretability, which is a requirement established by customs officers. In this paper, we propose a new Bayesian-based hybrid approach combining machine learning and multi-objective linear programming (MOLP), called multi-objective Bayesian with dynamic optimization (MOBADO). We demonstrate that it is possible to more than double (with a 237% increase) the precision of current inspection systems, freeing up almost 50% of human resources, and outperform past results with respect to each of the above objectives. MOBADO is an optimization technique that could be combined with any artificial intelligence approach capable of optimizing the quality of multi-objective risk analysis in real time. [ABSTRACT FROM AUTHOR]

Gender violence is a problem that affects millions of people worldwide. Among its many manifestations Intimate Partner Violence (IPV) is one of the most common. In Spain, a police monitoring protocol has been developed to minimize recidivism in IPV cases. This protocol is complemented by VioGén, an Intimate Partner Violence Risk Assessment Tool (IPVRAT) created by the Spanish State Secretariat for Security of the Ministry of Interior (SES) for risk prediction. VioGén's goal is to help the authorities determine what security and safety measures are most suitable. This paper improves on the current version of VioGén by introducing a model based on machine learning and data science and by studying the predictive value of exogenous and historical variables. The model is fitted on an anonymized database provided by SES and extracted from VioGén. This database includes the 2-year evolution of 46,047 new cases of IPV violence reported between October 2016 and December 2017, making it the largest database analyzed in the field. Obtained results show a clear improvement in the predictive capabilities of the new model against the original system, where it would have corrected more than 25% of the infra-protected cases, while improving the overall accuracy at the same time. Finally, lessons learned from the performed study and experiments are reported to aid in the design of future IPVRAT. In particular, insights show that IPVRAT should not treat cases statically as the incorporation of information regarding their evolution improves significantly the model's performance. • Analysis of the largest dataset in IPVRAT to date: 2-year evolution of 46,047 cases. • Study of the predictive value of exogenous and historical variables. • New research paradigm: the direct estimation of the Optimal Protection Level. • Comparison of multiclass vs ordinal classification paradigms in the context of IPV. • New model that corrects over 25% of the cases infra-protected by the original system. [ABSTRACT FROM AUTHOR]

• Predictive modelling benchmark of Nitrate Vulnerable Zones using Random Forest. • Extrinsic driving forces to groundwater were selected as environmental predictors. • Phenological features derived from remote sensing were included as novel features. • Feature selection methods revealed good performance predicting nitrate pollution. • Phenology and manure production from livestock farms as most important features. Nitrate leaching losses from arable lands into groundwater were a main driver in designating Nitrate Vulnerable Zones (NVZs) according to the Nitrates Directive, with a view to enhancing their water quality. Despite this, developing common strategies for effective water quality control in these areas remains a challenge in the European Union. This paper evaluates the performance of the Random Forest (RF) machine learning algorithm combined with Feature Selection (FS) techniques in predicting nitrate pollution in NVZs groundwater bodies in different periods and using updated environmental features in Andalusia, Spain. A set of forty-four features extrinsic to groundwater bodies were used as environmental predictors, with an aim to make this methodology exportable to other regions. Phenological features obtained through remote-sensing techniques were included to measure the dynamics of agricultural activity. In addition, other dynamic features derived from weather and livestock effluents were included to analyse seasonal and interannual changes in nitrate pollution. Three feature stacks and two nitrate databases were used in the predictive modelling: Period 1 (2009), with 321 nitrate samples for training; Period 2 (2010), with 282 nitrate samples for validation and initial spatial prediction; and Period 3 (2017), to assess the changes in the probability of groundwater nitrate content exceeding 50 mg/L. Random Forest as a wrapper with four sequential search methods was considered: sequential backward selection (SBS), sequential forward selection (SFS), sequential forward floating selection (SFFS) and sequential backward floating selection (SBFS). From among all the Feature Selection methods applied, Random Forest with SFS had the best performance (overall accuracy = 0.891 and six predictor features) and linked the highest probability of nitrate pollution with three dynamic features: the Normalized Difference Vegetation Index (NDVI) base level, NDVI value for the end of the growing season and accumulated manure production of livestock farms; and three static features: slope, sediment depositional areas and valley depth. [ABSTRACT FROM AUTHOR]

• Groundwater-dependent wetlands rank among the world's most endangered ecosystems. • A novel machine learning approach to map groundwater-dependent ecosystems is presented. • Multi-parametric supervised classification rendered test scores over 96%. • Support vector machines, tree-based classifiers and logistic regression proved suitable. • Results could be used to inform wetland management and land use policy. Identifying groundwater-dependent ecosystems is the first step towards their protection. This paper presents a machine learning approach that maps groundwater-dependent ecosystems by extrapolating from the characteristics of a small sample of known wetland and non-wetland areas to find other areas with similar geological, hydrological and biotic markers. Explanatory variables for wetland occurrence include topographic elevation, lithology, vegetation vigor, and slope-related variables, among others. Supervised classification algorithms are trained based on the ground truth sample, and their outcomes are checked against an official inventory of groundwater-dependent ecosystems for calibration. This method is illustrated through its application to a UNESCO Biosphere Reserve in central Spain. Support vector machines, tree-based classifiers, logistic regression and k-neighbors classification predicted the presence of groundwater-dependent ecosystems adequately (>96% test and AUC scores). The ensemble mean of the best five classifiers rendered a 90% success rate when computed per surface area. This method can optimize fieldwork during the characterization stage of groundwater-dependent ecosystems, thus contributing to integrate wetland protection in land use planning. [ABSTRACT FROM AUTHOR]

[Display omitted] • A novel genetic algorithm-based seasonal fractional grey model is proposed. • Monthly electricity generation is forecasted since the Covid-19 lockdown started. • Seasonal grey and machine learning models are used for forecasting. • The selected countries are France, Germany, Spain, Turkey, and the UK. • The share of renewables in total electricity generation is forecasted. Balances in the energy sector have changed since the implementation of the Covid-19 pandemic lockdown in Europe. This paper analyses how the lockdown affected electricity generation in European countries and how it will reshape future energy generation. Monthly electricity generation from total renewables and non-renewables in France, Germany, Spain, Turkey, and the UK from January 2017 to September 2020 were evaluated and compared. Four seasonal grey prediction models and three machine learning methods were used for forecasting; the quarterly results are presented to the end of 2021. Additionally, the share of electricity generation from renewables in total electricity generation from 2017 to 2021 for the selected countries was compared. Electricity generation from total non-renewables in the second quarter of 2020 for France, Germany, Spain, and the UK decreased by 21%–25% compared to the same period of 2019; the decline in Turkey was approximately 11%. Additionally, electricity generation from non-renewables in the third quarter of 2020 for all countries, except Turkey, decreased compared to the same period of the previous year. All grey prediction models and support vector machine method forecast that the share of renewables in total electricity generation will increase continuously in France, Germany, Spain, and the UK to the end of 2021. The forecasting methods provided by this study open new avenues for research on the impact of the Covid-19 pandemic on the future of the energy sector. [ABSTRACT FROM AUTHOR]

• Harmonization and quality control of originally heterogenous observational databases. • Large-scale application of machine learning methods in a distributed data network. • Transparent use of open-source software tools and publicly shared analytical code. As a response to the ongoing COVID-19 pandemic, several prediction models in the existing literature were rapidly developed, with the aim of providing evidence-based guidance. However, none of these COVID-19 prediction models have been found to be reliable. Models are commonly assessed to have a risk of bias, often due to insufficient reporting, use of non-representative data, and lack of large-scale external validation. In this paper, we present the Observational Health Data Sciences and Informatics (OHDSI) analytics pipeline for patient-level prediction modeling as a standardized approach for rapid yet reliable development and validation of prediction models. We demonstrate how our analytics pipeline and open-source software tools can be used to answer important prediction questions while limiting potential causes of bias (e.g. , by validating phenotypes, specifying the target population, performing large-scale external validation, and publicly providing all analytical source code). We show step-by-step how to implement the analytics pipeline for the question: 'In patients hospitalized with COVID-19, what is the risk of death 0 to 30 days after hospitalization?'. We develop models using six different machine learning methods in a USA claims database containing over 20,000 COVID-19 hospitalizations and externally validate the models using data containing over 45,000 COVID-19 hospitalizations from South Korea, Spain, and the USA. Our open-source software tools enabled us to efficiently go end-to-end from problem design to reliable Model Development and evaluation. When predicting death in patients hospitalized with COVID-19, AdaBoost, random forest, gradient boosting machine, and decision tree yielded similar or lower internal and external validation discrimination performance compared to L1-regularized logistic regression, whereas the MLP neural network consistently resulted in lower discrimination. L1-regularized logistic regression models were well calibrated. Our results show that following the OHDSI analytics pipeline for patient-level prediction modelling can enable the rapid development towards reliable prediction models. The OHDSI software tools and pipeline are open source and available to researchers from all around the world. [ABSTRACT FROM AUTHOR]

Abstract: Land cover monitoring using remotely sensed data requires robust classification methods which allow for the accurate mapping of complex land cover and land use categories. Random forest (RF) is a powerful machine learning classifier that is relatively unknown in land remote sensing and has not been evaluated thoroughly by the remote sensing community compared to more conventional pattern recognition techniques. Key advantages of RF include: their non-parametric nature; high classification accuracy; and capability to determine variable importance. However, the split rules for classification are unknown, therefore RF can be considered to be black box type classifier. RF provides an algorithm for estimating missing values; and flexibility to perform several types of data analysis, including regression, classification, survival analysis, and unsupervised learning. In this paper, the performance of the RF classifier for land cover classification of a complex area is explored. Evaluation was based on several criteria: mapping accuracy, sensitivity to data set size and noise. Landsat-5 Thematic Mapper data captured in European spring and summer were used with auxiliary variables derived from a digital terrain model to classify 14 different land categories in the south of Spain. Results show that the RF algorithm yields accurate land cover classifications, with 92% overall accuracy and a Kappa index of 0.92. RF is robust to training data reduction and noise because significant differences in kappa values were only observed for data reduction and noise addition values greater than 50 and 20%, respectively. Additionally, variables that RF identified as most important for classifying land cover coincided with expectations. A McNemar test indicates an overall better performance of the random forest model over a single decision tree at the 0.00001 significance level. [Copyright &y& Elsevier]

Abstract: Breast screening is a method of detecting breast cancer at a very early stage. The first step involves taking an X-ray, called a mammogram, of each breast. The mammogram can detect small changes in breast tissue which may indicate cancers which are too small to be felt either by the woman herself or by a doctor. The World Health Organisation’s International Agency for Research on Cancer (IARC) concluded that mammography screening for breast cancer reduces mortality. This means that out of every 500 women screened, one life will be saved. The present research uses the information obtained from the breast screening programme carried out in the public health area of Aviles (Principality of Asturias, Spain) from 1999 to 2007. The public health area of Aviles is formed by nine municipalities with a total of 160,000 inhabitants. The selection of the public health area was based on the following criteria: [•] This is the first screening programme performed in the area. [•] Almost 100% of the population in the area benefit from the public health system. [•] The Aviles public health area is a well-defined area of the region that does not send patients to other public health areas, which makes the study easier and more accurate. This paper describes a neural network based approach to breast cancer diagnosis; the model developed is able to determine which women are more likely to suffer from a particular kind of tumour before they undergo a mammography. [ABSTRACT FROM AUTHOR]

Thermostatic load control systems are widespread in many countries. Since they provide heat for domestic hot water and space heating on a massive scale in the residential sector, the assessment of their energy performance and the effect of different control strategies requires simplified modeling techniques demanding a small number of inputs and low computational resources. Data-driven techniques are envisaged as one of the best options to meet these constraints. This paper presents a novel methodology consisting of the combination of an optimization algorithm, two auto-regressive models and a control loop algorithm able to virtually replicate the control of thermostatically driven systems. This combined strategy includes all the thermostatically controlled modes governed by the set point temperature and enables automatic assessment of the energy consumption impact of multiple scenarios. The required inputs are limited to available historical readings from smart thermostats and external climate data sources. The methodology has been trained and validated with data sets coming from a selection of 11 smart thermostats, connected to gas boilers, placed in several households located in north-eastern Spain. Important conclusions of the research are that these techniques can estimate the temperature decay of households when the space heating is off as well as the energy consumption needed to reach the comfort conditions. The results of the research also show that estimated median energy savings of 18.1% and 36.5% can be achieved if the usual set point temperature schedule is lowered by 1 °C and 2 °C, respectively. [ABSTRACT FROM AUTHOR]

Support vector machines (SVM) represent a promising development in machine learning research that is not widely used within the remote sensing community. This paper reports the results of two experiments in which multi-class SVMs are compared with maximum likelihood (ML) and artificial neural network (ANN) methods in terms of classification accuracy. The two land cover classification experiments use multispectral (Landsat-7 ETM+) and hyperspectral (DAIS) data, respectively, for test areas in eastern England and central Spain. Our results show that the SVM achieves a higher level of classification accuracy than either the ML or the ANN classifier, and that the SVM can be used with small training datasets and high-dimensional data. [ABSTRACT FROM AUTHOR]

This paper summarises the development of a machine-readable model series for explaining Gaudı´'s use of ruled-surface geometry in the Sagrada Familia in Barcelona, Spain. The first part discusses the modeling methods underlying the columns of the cathedral and the techniques required to translate them into built structures. The second part discusses the design and development of a tangible machine-readable model to explain column-modeling methods interactively in educational contexts such as art exhibitions. It is designed to explain the principles underlying the column design by means of physical interaction without using mathematical terms or language. [Copyright &y& Elsevier]

This work presents a full statistical analysis and accurate prediction of low-visibility events due to fog, at the A-8 motor-road in Mondoñedo (Galicia, Spain). The present analysis covers two years of study, considering visibility time series and exogenous variables collected in the zone affected the most by extreme low-visibility events. This paper has then a two-fold objective: first, we carry out a statistical analysis for estimating the fittest probability distributions to the fog event duration, using the Maximum Likelihood method and an alternative method known as the L-moments method. This statistical study allows association of the low-visibility depth with the event duration, showing a clear relationship, which can be modeled with distributions for extremes such as Generalized Extreme Value and Generalized Pareto distributions. Second, we apply a neural network approach, trained by means of the ELM (Extreme Learning Machine) algorithm, to predict the occurrence of low-visibility events due to fog, from atmospheric predictive variables. This study provides a full characterization of fog events at this motor-road, in which orographic fog is predominant, causing important traffic problems during all year. We also show how the ELM approach is able to obtain highly accurate low-visibility events predictions, with a Pearson correlation coefficient of 0.8 , within a half-hour time horizon, enough to initialize some protocols aiming at reducing the impact of these extreme events in the traffic of the A-8 motor road. [ABSTRACT FROM AUTHOR]

Currently, there is a lack of accurate simulation tools for the thermal performance modeling of courtyards due to their intricate thermodynamics. Machine Learning (ML) models have previously been used to predict and evaluate the structural performance of buildings as a means of solving complex mathematical problems. Nevertheless, the microclimatic conditions of the building surroundings have not been as thoroughly addressed by these methodologies. To this end, in this paper, the adaptation of ML techniques as a more comprehensive methodology to fill this research gap, covering not only the prediction of the courtyard microclimate but also the interpretation of experimental data and pattern recognition, is proposed. Accordingly, based on the climate zoning and aspect ratios of 32 monitored case studies located in the South of Spain, the Support Vector Regression (SVR) method was applied to predict the measured temperature inside the courtyard. The results provided by this strategy showed good accuracy when compared to monitored data. In particular, for two representative case studies, if the daytime slot with the highest urban overheating is considered, the relative error is almost below 0.05%. Additionally, values for statistical parameters are in good agreement with other studies in the literature, which use more computationally expensive CFD models and show more accuracy than existing commercial tools. [ABSTRACT FROM AUTHOR]