ABSTRACT The growing reliance on fossil fuels for energy generation has raised concerns about their significant contribution to global warming and the associated risks of supply instability. Anaerobic Digestion (AD) within Wastewater Treatment Plants (WWTPs) offers a renewable alternative by producing biogas, while effective operational optimisation requires accurate forecasting of biogas yields under varying conditions. This study addresses this challenge by developing, tuning and evaluating five Deep Learning (DL) architectures for biogas production prediction: one‐dimensional Convolutional Neural Network (1D‐CNN), Long Short‐Term Memory (LSTM), Gated Recurrent Unit (GRU), Transformers and Residual Encoding. Among these, the GRU model demonstrated superior performance, achieving a Root Mean Square Error (RMSE) of 139.1 m 3 /day and a Mean Absolute Error (MAE) of 135.9 m 3 /day. The adaptability of the GRU model to different datasets was examined through Transfer Learning (TL), revealing a clear difference in performance depending on the TL approach used: the retrained model achieved a RMSE of 230.1 m 3 /day and a MAE of 229.9 m 3 /day, whereas the model without retraining exhibited higher errors of 358.7 and 358.8 m 3 /day, respectively. A key contribution of this work lies in its comprehensive Explainable Artificial Intelligence (XAI) analysis, which applied both ante hoc attention mechanisms and post hoc interpretability techniques such as SHAP and LIME. The XAI methods consistently identified biogas production, the study's target variable, as the most influential feature in the model's predictions. Among the remaining features, some changes were observed in their impact on model predictions. Moreover, the study highlighted how TL affects prediction performance and the stability and consistency of feature importance, thereby improving the transparency and trustworthiness of the forecasting models.

This article is a multidisciplinary and exhaustive commentary on Art 18 of Directive (EU) 2023/2225, of October 18, 2025 (CCD2), mainly focusing on Consumer Creditworthiness assessment (CWA) based on Artificial Intelligence (AI). Methodologically, we combine normative analysis with explanatory engineering and qualitative evidence from cognitive psychology. Beginning on the fundamental implications of the regime, through a text-based case study, also illustrated using a hypothetical Artificial Neural Network (ANN), we examine the scope and limits of the right to explanation enshrined in Art. 18(8)(a)(c). The analysis compares three main categories of explanations - global, local, and counterfactual - and shows that only the latter, although not sufficient, come close to a true action-oriented and future-proof duty of automated justifications, Also as our contribution, we propose overcoming mandatory human supervision in contexts where AI systems prove to be systematically superior to human performance, defending Human-Out-of-the-Loop (HOOTL) as a guiding principle in those circumstances. This investigation thus proposes a more pragmatic regulatory model that combines legal requirements, cognitive limitations, and existing and forthcoming technical capabilities in AI-based CWA.

With the increasing adoption of AI in safety-critical applications within urban environments, the interpretability of these systems is paramount. This study explores the application of Explainable Artificial Intelligence (XAI) techniques to enhance transparency in audio-based detection of emergency vehicle sirens, a crucial component in urban sound management. Adopting methods such as SHAP (SHapley Additive exPlanations) values, Permutation Feature Importance, and model-specific feature scores, this research identifies key audio features, including mid-frequency spectral contrasts and targeted chroma components, which significantly help in distinguishing siren sounds among urban noise. The study examines various machine learning models, identifying K-Nearest Neighbors (KNN) and XGBoost as top performers; KNN excelled in class-specific precision, while XGBoost demonstrated strong cross-class discrimination. The findings highlight the potential of XAI in improving both accuracy and accountability for sound detection systems in safety-critical urban applications, advancing the deployment of transparent AI within smart city infrastructures.

Predictive maintenance (PdM) in Industry 4.0 (I4.0) increasingly relies on machine learning (ML) techniques to minimize unplanned downtime and enhance operational efficiency. While cloud-based ML solutions offer scalability and strong predictive performance, their reliance on network connectivity introduces latency and reliability issues that hinder real-time industrial applications. This study investigates the deployment of lightweight autoencoder (AE)-based models optimized for edge computing environments, comparing their performance against traditional cloud-hosted alternatives. Multiple model architectures were evaluated, and inference latency was benchmarked across four deployment scenarios: cloud-hosted PyTorch, native PyTorch on Raspberry Pi 3B, TensorFlow Lite (Python runtime), and TensorFlow Lite (C++ runtime). Latency measurements, averaged 100 executions per model, reveal that edge deployment can reduce inference time by up to 7000× compared to cloud containers, with TensorFlow Lite C++ deployments achieving latencies as low as 60 microseconds. These results demonstrate that edge-based ML deployment is a viable strategy for enabling timely, autonomous fault detection in real-time PdM systems.

Abstract Violence recognition is challenging because it can be presented in very different forms. For example, it can be present in an image by a person hitting another person or present in audio by a person being rude to another. Thus, audio and video are essential features to be analysed. In the audio approach, speech processing, music, and ambient sound are some of the main points of this problem since finding similarities and differences between these domains is necessary. Human activity can be classified into four different categories in the video approach, depending on the complexity and the number of body parts involved in the action. Examples of Human activity categories are considered: gestures, actions, interactions and activities. Recognizing human actions in the video becomes a challenge with this varied set of human activities. Furthermore, in the last years, the growth of deep learning techniques applied to this area has been enormous, and the reason is that their results surpass traditional signal processing on a large scale. This article is based on audio and video signals inside a vehicle to detect violence. Furthermore, the architecture used was ResNet model with Mel‐spectrogram methodology for audio signals. The proposed method for video signal representation was RGB, which applied four different models: C2D, I3D, X3D, and Flow‐Gated. Finally, multimodal fusion was applied at the end of the process.

Population growth has increased the demand for meat and dairy products, making livestock, especially cattle, key to meeting this demand. This has led to an increase in herd size, complicating efficient herd management. To meet this challenge, innovative technologies, such as monitoring collars, have been developed to improve individual animal management. This research work evaluates and compares three unsupervised anomaly detection methods to identify estrus in dairy cows from intensive farms, based on daily activity data recorded by a commercial monitoring collar. Data from two different dairy farms have been used and the results have been compared by evaluating the behavior both individually and at herd level. The results obtained show a good performance of the selected techniques in the individual animal models. Thus, this research demonstrates that these techniques can be very useful tools in farm management, providing valuable information, improving productivity and, consequently, increasing the economic performance of the farm.

Abstract In recent years, we have seen a growing need for energy, which has had environmental consequences through the use of fossil fuels. Some of the sectors of our society make intensive use of energy, as is the case with wastewater treatment plants (WWTPs). Through anaerobic digestion, these infrastructures can produce energy, therefore improve energy efficiency and decrease the environmental footprint. This study aims to design, tune and evaluate a hybrid deep learning (DL) model, called incremental dual path network (IDPN), to forecast energy production in an anaerobic reactor for the next two days. The hybrid model’s performance was compared against five DL models conceived: long short-term memory (LSTMs), multi-layer perception (MLP), gated recurrent units (GRUs), Transformers and convolutional neural networks (CNNs). Furthermore, two data processing strategies were applied due to system failures and missing values. Four model evaluation metrics and the obtained results show that the hybrid model, which combines LSTMs and CNNs, presented the best performance in both approaches of data processing, with the best candidate model presenting a mean absolute error (MAE) of 312.1 kWh, root mean squared error (RMSE) of 341.6 kWh, mean absolute percentage error (MAPE) of 15.9% and R $$^{2}$$ 2 of 0.95. Following this, an ablation study was conducted, demonstrating that across several variations, the baseline IDPN consistently achieved the best results. Moreover, in both approaches, the removal or modification of the CNN led to a severe decline in performance, surpassing the impact of altering the LSTM, reinforcing its importance in the model’s architecture.

Abstract Object detection is one of the most fundamental problems to tackle in the computer vision research area. Recent advances in multimodal data streams and deep learning architectures have prompted a fast growth in the field of multimodal learning, which brings several advantages over single-modality approaches for object detection, such as improved accuracy, robustness to noise and ambiguity, handling of complex scenarios and adaptability to diverse data. Some of the biggest challenges when implementing a multimodal learning approach are the selection of the fusion strategy, design of processing architecture, modality alignment/synchronization and interpretability of such high-dimensional representations. To address this challenge, we propose a feature-level fusion architecture for object detection based on extracting YOLO features from images, spectral and rhythm features from sound using Mel-frequency cepstral coefficients, and general descriptors from radar modalities that, after timestamp and homography transformation matrix alignment, are combined with an attention mechanism into a single classification network. Preliminary experiments indicate that the proposed architecture can constitute itself as a base pipeline for several different multimodal object detection tasks in real-world applications.

In recent years, in the field of education, there has been a progressive trend towards teaching that is more personalised to students' characteristics and some models of prediction failure that are more accurate. In this sense, machine learning techniques have contributed to this realisation. This transformation has been significantly influenced by the integration of machine learning techniques, which have played a crucial role in harnessing data to enhance educational practices. This paper investigates the transformative potential of Artificial Intelligence (AI) within secondary education, focusing on the utilization of student assessment data and socioeconomic contextual information. The primary objective of this paper is to investigate the transformative potential of Artificial Intelligence (AI) within secondary education, with a specific focus on the utilization of student assessment data and socioeconomic contextual information. So, this paper explores the application of AI algorithms to create tailored learning pathways, adaptive support mechanisms, and targeted interventions that accommodate diverse student backgrounds. The integration of AI in secondary education is envisioned not only as a means to enhance academic outcomes but also as a tool to promote social equity and inclusivity. By leveraging data insights, educators can identify and respond to the unique needs of each student, fostering an environment where learning is optimized for individual growth. Furthermore, the paper scrutinizes the ethical considerations and challenges inherent in deploying AI systems in educational settings, emphasizing the pivotal role of equity, transparency, and data privacy in these implementations. This research aims to offer educators, policymakers, and stakeholder's insights into harnessing AI to foster adaptable, student-centric learning environments that bridge educational gaps and promote holistic academic development in secondary schools. Ethical guidelines and frameworks are discussed to ensure responsible AI deployment in educational contexts, safeguarding the rights and privacy of students. The data explored is taken from the management system of a secondary school in the municipality of Braga, relating to students taking Maths A, Maths B or Maths Applied to the Social Sciences (MACS). A total of 621 students were analysed, of which: 520 students attended Mathematics A in the Science and Technology and Socio-Economic Sciences courses, 20 attended Mathematics B in the Visual Arts course and 81 attended Mathematics Applied to Social Sciences in the Languages and Humanities course. The different maths subjects were analysed separately and at the end a comparative study was carried out between the three strands. Grounded in the analysis of these integrated datasets, the study sheds light on the pivotal role of AI in revolutionizing secondary school education. By closely examining student assessment data alongside socioeconomic indicators, such as academic performance and behavioral patterns, the paper identifies opportunities for AI integration to personalize learning experiences, address educational disparities, and cultivate inclusive learning environments. In Maths A there are 268 female and 252 male students. In Maths B there are 18 females and 2 males (most of the arts subjects are taken by women). In MACS the distribution is as follows: 45 female and 36 male. All these students, regardless of the area they chose at the start of secondary school, have in common the choice of Maths, however Maths A will be taken throughout secondary school, while the other two strands are only taken in the first two years of secondary school. Data fusuion techniques have achieved better performance in this type of study, since the models themselves combine the predictions of two or more base models.

Abstract One of the most common attacks is man‐in‐the‐middle (MitM) which, due to its complex behaviour, is difficult to detect by traditional cyber‐attack detection systems. MitM attacks on internet of things systems take advantage of special features of the protocols and cause system disruptions, making them invisible to legitimate elements. In this work, an intrusion detection system (IDS), where intelligent models can be deployed, is the approach to detect this type of attack considering network alterations. Therefore, this paper presents a novel method to develop the intelligent model used by the IDS, being this method based on a hybrid process. The first stage of the process implements a feature extraction method, while the second one applies different supervised classification techniques, both over a message queuing telemetry transport (MQTT) dataset compiled by authors in previous works. The contribution shows excellent performance for any compared classification methods. Likewise, the best results are obtained using the method with the highest computational cost. Thanks to this, a functional IDS will be able to prevent MQTT attacks.

The complexity associated with groups of tourists led to the emergence of Group Recommender Systems (GRS) for tourism. But if generating recommendations for small groups is a complex task, to provide them to large and occasional groups is even more. This complexity is especially due to the group’s heterogeinity, conflicting preferences, the information overload found on the internet and the tourists’ different ways of coping with the information, hindering the recommendation process from the users’ profile construction to the final recommendation of a list of points of interest to visit. In this work, we show how we tackled the identified issues in a GRS prototype, Grouplanner, including the cold-start problem, by predicting the tourists’ preferences based only on their personality and dividing the main group into subgroups of similar personality; by using a Multi-Agent Microservice; a novel dynamic clustering algorithm, d-means, adapted from the k-means algorithm, that does not need to know the number of clusters a priori; and association rules. Using a personality dataset of n=100k users, the proposed d-means algorithm was tested against two baselines (k-means and k-means++), showing better results in the clustering quality and scalability. We were also able to determine a large set of association rules to refine the recommendations, although further improvements are needed. To test the Grouplanner prototype, a simulation with real users (n=35) was conducted. The results showed the subgroups formed were very compact, revealing a very good clustering quality, with an average silhouette of s = 0.91. 11 of the 15 proposed tourist preferences were successfully predicted and used for the preliminary recommendation lists, being 92 % of the participants satisfied with the individual recommendations and 96 % with the group recommendations.

The complexity associated to group recommendations needs strategies to mitigate several problems, such as the group's heterogeinity and conflicting preferences, the emotional contagion phenomenon, the cold-start problem, and the group members’ needs and concerns while providing recommendations that satisfy all members at once. In this demonstration, we show how we implemented a Multi-Agent Microservice to model the tourists in a mobile Group Recommender System for Tourism prototype and a novel dynamic clustering process to help minimize the group's heterogeneity and conflicting preferences. To help solve the cold-start problem, the preliminary tourist attractions preference and travel-related preferences & concerns are predicted using the tourists' personality, considering the tourists’ disabilities and fears/phobias. Although there is no need for data from previous interactions to build the tourists’ profile since we predict the tourists’ preferences, the tourist agents learn with each other by using association rules to find patterns in the tourists' profile and in the ratings given to Points of Interest to refine the recommendations.

Abstract Machine learning has been facing significant challenges over the last years, much of which stem from the new characteristics of machine learning problems, such as learning from streaming data or incorporating human feedback into existing datasets and models. In these dynamic scenarios, data change over time and models must adapt. However, new data do not necessarily mean new patterns. The main goal of this paper is to devise a method to predict a model's performance metrics before it is trained, in order to decide whether it is worth it to train it or not. That is, will the model hold significantly better results than the current one? To address this issue, we propose the use of meta‐learning. Specifically, we evaluate two different meta‐models, one built for a specific machine learning problem, and another built based on many different problems, meant to be a generic meta‐model, applicable to virtually any problem. In this paper, we focus only on the prediction of the root mean square error (RMSE). Results show that it is possible to accurately predict the RMSE of future models, event in streaming scenarios. Moreover, results also show that it is possible to reduce the need for re‐training models between 60% and 98%, depending on the problem and on the threshold used.

A large percentage of the worldwide population is affected by chronic diseases, leading to a burden of the patient and the national healthcare systems. Recommendation systems are used for the personalization of healthcare due to their capacity of performing predictive analyses based on the patient’s clinical data. This systematic literature review presents four research questions to provide an overall state of the art of the use of recommendation systems applied to the healthcare of patients with chronic diseases. Disease management was identified as the main purpose of the systems proposed in the literature. However, few solutions provide support to physicians in the clinical decision-making. Ontologies and rule-based systems were the artificial intelligence techniques most used in the systems since they can easily implement clinical guidelines. Current challenges of these systems include the low adherence, data sparsity, heterogeneous data, and explainability, that affect the success of the recommendation system. The results also show that there are few systems that provide support to patients with multiple chronic conditions. The findings of this literature review should be considered in the development of future recommendation systems that aim to support the management of chronic diseases.

In the last years, the number of machine learning algorithms and their parameters has increased significantly. On the one hand, this increases the chances of finding better models. On the other hand, it increases the complexity of the task of training a model, as the search space expands significantly. As the size of datasets also grows, traditional approaches based on extensive search start to become prohibitively expensive in terms of computational resources and time, especially in data streaming scenarios. This paper describes an approach based on meta-learning that tackles two main challenges. The first is to predict key performance indicators of machine learning models. The second is to recommend the best algorithm/configuration for training a model for a given machine learning problem. When compared to a state-of-the-art method (AutoML), the proposed approach is up to 130x faster and only 4% worse in terms of average model quality. Hence, it is especially suited for scenarios in which models need to be updated regularly, such as in streaming scenarios with big data, in which some accuracy can be traded for a much shorter model training time.

Abstract To travel in leisure is an emotional experience, and therefore, the more the information about the tourist is known, the more the personalized recommendations of places and attractions can be made. But if to provide recommendations to a tourist is complex, to provide them to a group is even more. The emergence of personality computing and personality-aware recommender systems (RS) brought a new solution for the cold-start problem inherent to the conventional RS and can be the leverage needed to solve conflicting preferences in heterogenous groups and to make more precise and personalized recommendations to tourists, as it has been evidenced that personality is strongly related to preferences in many domains, including tourism. Although many studies on psychology of tourism can be found, not many predict the tourists’ preferences based on the Big Five personality dimensions. This work aims to find how personality relates to the choice of a wide range of tourist attractions, traveling motivations, and travel-related preferences and concerns, hoping to provide a solid base for researchers in the tourism RS area to automatically model tourists in the system without the need for tedious configurations, and solve the cold-start problem and conflicting preferences. By performing Exploratory and Confirmatory Factor Analysis on the data gathered from an online questionnaire, sent to Portuguese individuals from different areas of formation and age groups ( n = 1035), we show all five personality dimensions can help predict the choice of tourist attractions and travel-related preferences and concerns, and that only neuroticism and openness predict traveling motivations.

The identification of the emotional states corresponding to the four quadrants of the valence/arousal space has been widely analyzed in the scientific literature by means of multiple techniques. Nevertheless, most of these methods were based on the assessment of each brain region separately, without considering the possible interactions among different areas. In order to study these interconnections, this study computes for the first time the functional connectivity metric called cross-sample entropy for the analysis of the brain synchronization in four groups of emotions from electroencephalographic signals. Outcomes reported a strong synchronization in the interconnections among central, parietal and occipital areas, while the interactions between left frontal and temporal structures with the rest of brain regions presented the lowest coordination. These differences were statistically significant for the four groups of emotions. All emotions were simultaneously classified with a 95.43% of accuracy, overcoming the results reported in previous studies. Moreover, the differences between high and low levels of valence and arousal, taking into account the state of the counterpart dimension, also provided notable findings about the degree of synchronization in the brain within different emotional conditions and the possible implications of these outcomes from a psychophysiological point of view.

Abstract The devastating consequences of climate change have resulted in the promotion of clean energies, being the wind energy the one with greater potential. This technology has been developed in recent years following different strategic plans, playing special attention to wind generation. In this sense, the use of bicomponent materials in wind generator blades and housings is a widely spread procedure. However, the great complexity of the process followed to obtain this kind of materials hinders the problem of detecting anomalous situations in the plant, due to sensors or actuators malfunctions. This has a direct impact on the features of the final product, with the corresponding influence in the durability and wind generator performance. In this context, the present work proposes the use of a distributed anomaly detection system to identify the source of the wrong operation. With this aim, five different one-class techniques are considered to detect deviations in three plant components located in a bicomponent mixing machine installation: the flow meter, the pressure sensor and the pump speed.

Online video games drive a multi-billion dollar industry dedicated to maintaining a competitive and enjoyable experience for players. Traditional cheat detection systems struggle when facing new exploits or sophisticated fraudsters. More advanced solutions based on machine learning are more adaptive but rely heavily on in-game data, which means that each game has to develop its own cheat detection system. In this work, we propose a novel approach to cheat detection that doesn't require in-game data. Firstly, we treat the multimodal interactions between the player and the platform as multivariate time series. We then use convolutional neural networks to classify these time series as corresponding to legitimate or fraudulent gameplay. Our models achieve an average accuracy of respectively 99.2% and 98.9% in triggerbot and aimbot (two widespread cheats), in an experiment to validate the system's ability to detect cheating in players never seen before. Because this approach is based solely on player behavior, it can be applied to any game or input method, and even various tasks related to modeling human activity.

According to the World Health Organization (WHO), close to 300 million people of all ages suffer from depression. Also, for WHO, depression is the leading reason for disability worldwide and is a major contributor to the global burden of disease. Different than the mood fluctuation raised by the common life's activities, depression can be a serious health problem, particularly when it is a long-term and mid/high intensity. Luckily, despite depression is a silent disease, people when suffering leaves some clues. Due to the massive use of social media, these clues can be collected through the texts posted on social media, such as Twitter, Facebook, Instagram, and later, analysed to identify if the writing style matches with a depressive pattern. This paper presents an approach that can be applied by Machine Learning models to help psychologists to identify depressive clues in texts. The model examines profiles on Twitter based on clues provided by users in their posts. Combining Sentiment Analysis, Machine Learning and Natural Language Processing techniques, we achieved a precision of 98% by Machine Learning models when identifying Twitter profiles that post potential depressive texts.

Environmental sustainability is one of the biggest concerns nowadays. With increasingly latent negative impacts, it is substantiated that future generations may be compromised. The research here presented addresses this topic, focusing on air quality and atmospheric pollution, in particular the Ultraviolet index and Carbon Monoxide air concentration, as well as water issues regarding Wastewater Treatment Plants, in particular the pH of water. A set of Machine Learning regressors and classifiers are conceived, tuned, and evaluated in regard to their ability to forecast several parameters of interest. The experimented models include Decision Trees, Random Forests, Multilayer Perceptrons, and Long Short-Term Memory networks. The obtained results assert the strong ability of LSTMs to forecast air pollutants, with all models presenting similar results when the subject was the pH of water.

Summary One of the most important factors to determine the success of an organization is the quality of decisions made. Supporting a decision‐making process is a complex task, mainly when decision‐makers are dispersed. Group decision support systems (GDSSs) have been studied over the last decades with the goal of providing support to decision‐makers; however, their acceptance by organizations has been difficult. This happens mostly due to usability problems, loss of interaction between decision‐makers, and consequently, loss of information. In this work, we present a web‐based GDSS developed to support groups of decision‐makers, regardless of their geographic location. The system allows the creation of multicriteria problems and the configuration of the preferences, intentions, and interests of each decision‐maker. The presented system uses a multiagent system to combine and process this information, using virtual agents that represent each decision‐maker. We believe that, with this approach, we will proceed in the refinements of a successful GDSS to correctly support decision‐makers while preserving the valuable intelligence and knowledge that can be generated in face‐to‐face meetings. Furthermore, the high level of usability that the system provides will contribute to an easier acceptance and adoption of this kind of systems.

Abstract Writing style is considered the manner in which an author expresses his thoughts, influenced by language characteristics, period, school, or nation. Often, this writing style can identify the author. One of the most famous examples comes from 1914 in Portuguese literature. With Fernando Pessoa and his heteronyms Alberto Caeiro, Álvaro de Campos, and Ricardo Reis, who had completely different writing styles, led people to believe that they were different individuals. Currently, the discussion of authorship identification is more relevant because of the considerable amount of widespread fake news in social media, in which it is hard to identify who authored a text and even a simple quote can impact the public image of an author, especially if these texts or quotes are from politicians. This paper presents a process to analyse the emotion contained in social media messages such as Facebook to identify the author's emotional profile and use it to improve the ability to predict the author of the message. Using preprocessing techniques, lexicon‐based approaches, and machine learning, we achieved an authorship identification improvement of approximately 5% in the whole dataset and more than 50% in specific authors when considering the emotional profile on the writing style, thus increasing the ability to identify the author of a text by considering only the author's emotional profile, previously detected from prior texts.

Abstract The need for integration of clinical practice guidelines (CPGs) in daily clinical practice calls for computational systems able to operationalise their knowledge and provide an enhanced experience in their enactment. Current approaches lack in functionalities such as scheduling and temporal management of CPGs, the combination of CPGs, and user‐friendly systems for computer‐interpretable guideline (CIG) creation and editing. This paper presents a comprehensive architecture for the deployment of CIGs, featuring components that allow the following: the creation and manipulation of clinical practice guideline knowledge elements, execution of CIGs with the temporal verification of clinical tasks, and drug conflict identification and resolution. This comprehensive approach provides a step‐by‐step assistant for health care professionals in the form of an agenda of activities that detects drug interactions when they are prescribed simultaneously and applies a mitigation algorithm to select possible and conflict‐free alternatives. This work addresses the lack of a unified pipeline and mitigation features shown in approaches to CIG conflict mitigation in use.

In the last years, organizations and companies in general have found the true potential value of collecting and using data for supporting decision-making. As a consequence, data are being collected at an unprecedented rate. This poses several challenges, including, for example, regarding the storage and processing of these data. Machine Learning (ML) is also not an exception, in the sense that algorithms must now deal with novel challenges, such as learn from streaming data or deal with concept drift. ML engineers also have a harder task when it comes to selecting the most appropriate model, given the wealth of algorithms and possible configurations that exist nowadays. At the same time, training time is a stronger restriction as the computational complexity of the training model increases. In this paper we propose a framework for dealing with these challenges, based on meta-learning. Specifically, we tackle two well-defined problems: automatic algorithm selection and continuous algorithm updates that do not require the retraining of the whole algorithm to adapt to new data. Results show that the proposed framework can contribute to ameliorate the identified issues.

Personalization is increasingly being perceived as an important factor for the effectiveness of Recommender Systems (RS). This is especially true in the tourism domain, where travelling comprises emotionally charged experiences, and therefore, the more about the tourist is known, better recommendations can be made. The inclusion of psychological aspects to generate recommendations, such as personality, is a growing trend in RS and they are being studied to provide more personalized approaches. However, although many studies on the psychology of tourism exist, studies on the prediction of tourist preferences based on their personality are limited. Therefore, we undertook a large-scale study in order to determine how the Big Five personality dimensions influence tourists' preferences for tourist attractions, gathering data from an online questionnaire, sent to Portuguese individuals from the academic sector and their respective relatives/friends (n=508). Using Exploratory and Confirmatory Factor Analysis, we extracted 11 main categories of tourist attractions and analyzed which personality dimensions were predictors (or not) of preferences for those tourist attractions. As a result, we propose the first model that relates the five personality dimensions with preferences for tourist attractions, which intends to offer a base for researchers of RS for tourism to automatically model tourist preferences based on their personality.

The present article proposes the training, testing and comparison of two models for ball detection, taking into account its final implementation in a Boccia game analysis computer-vision algorithm, within the “iBoccia” framework. The goal is to have a versatile and flexible algorithm towards different game environments. The selected ball detectors were a Histogram-of-Oriented-Gradients feature based Support Vector Machine (HOG-SVM) and a Convolutional Neural Network (CNN) based on a less complex implementation of the You Only Look Once model (Tiny-YOLO). Both detectors were evaluated offline and in real-time. The subsequent results showed that their performance was similar in both evaluations, however, Tiny-YOLO outperformed HOG-SVM by a small margin in all the used metrics. In real-time, both detectors achieved an accuracy of approximately 90%. Despite the high accuracy values, the detector requires further improvement because a single non-detection can influence the computer-vision algorithm's output, making the system unreliable.

Abstract Supporting group decision‐making when the decision makers are spread around the world is a complex process. The mechanisms of automated negotiation, such as argumentation, can be used in Ubiquitous Group Decision Support Systems (UbiGDSS) to help decision makers find a solution based on their preferences. However, the decision‐making process is much more than just a simple criteria and alternative analysis. There are many cognitive and affective issues that affect the outcome, and these issues should not be ignored; otherwise, the quality of the decision could be compromised. In this paper, we detail an UbiGDSS architecture and explore 2 cognitive and affective methods that are essential to the group decision‐making process. We explain how agents can reason about self‐expertise and other decision makers' credibility, and how agents can verify and react to tendencies throughout the decision‐making process. We intend agents to achieve higher quality and more consensual decisions. In any simulation environment that we tested, agents that analysed credibility, expertise, and/or analysed tendencies always achieved a higher consensus compared to agents that used neither of the proposed methods. Likewise, agents that used neither of the proposed methods or only performed tendencies analysis obtained the worst average satisfaction levels for each simulation environment.

Abstract Computer‐Interpretable Guidelines (CIGs) are the dominant medium for the delivery of clinical decision support, given the evidence‐based nature of their source material. Therefore, these machine‐readable versions have the ability to improve practitioner performance and conformance to standards, with availability at the point and time of care. The formalisation of Clinical Practice Guideline knowledge in a machine‐readable format is a crucial task to make it suitable for the integration in Clinical Decision Support Systems. However, the current tools for this purpose reveal shortcomings with respect to their ease of use and the support offered during CIG acquisition and editing. In this work, we characterise the current landscape of CIG acquisition tools based on the properties of guideline visualisation, organisation, simplicity, automation, manipulation of knowledge elements, and guideline storage and dissemination. Additionally, we describe the CompGuide Editor, a tool for the acquisition of CIGs in the CompGuide model for Clinical Practice Guidelines that also allows the editing of previously encoded guidelines. The Editor guides the users throughout the process of guideline encoding and does not require proficiency in any programming language. The features of the CIG encoding process are revealed through a comparison with already established tools for CIG acquisition.

The topic of stress is nowadays a very important one, not only in research but on social life in general. People are increasingly aware of this problem and its consequences at several levels: health, social life, work, quality of life, etc. This resulted in a significant increase in the search for devices and applications to measure and manage stress in real-time. Recent technological and scientific evolution fosters this interest with the development of new methods and approaches. In this paper we survey these new methods for stress assessment, focusing especially on those that are suited for the workplace: one of today's major sources of stress. We contrast them with more traditional methods and compare them between themselves, evaluating nine characteristics. Given the diversity of methods that exist nowadays, this work facilitates the stakeholders' decision towards which one to use, based on how much their organization values aspects such as privacy, accuracy, cost-effectiveness or intrusiveness.

Distributed Systems current face new challenges of adapting and reusing research results in the area of Intelligent Systems. Intelligent Systems use methods and technology derived from Knowledge-based and Computational Intelligence. Distributed Computing develops methods and technology to build systems composed of collaborating components. The fast growth of both Big Data and Data Mining have created interesting challenges for classical methods, algorithms, and frameworks from Distributed Computing, which makes especially interesting analysis and research into new trends and innovations that have recently appeared in this area. This special issue welcomed submissions of original papers introducing research results on all the aspects covering the roles of Knowledge and Intelligence in Distributed Systems, ranging from concepts and theoretical developments to advanced technologies and innovative applications. This issue presents a expanded versions of these papers from the best of those presented at the 10th International Symposium on Intelligent Distributed Computing (IDC 2016), which was held in Paris (France). As the special issue editors, we would like to take this opportunity to thank the various authors for their papers and the reviewers for their work. We are also grateful to Jon Hall, Editor-in-Chief of the Wiley journal Expert Systems. We would like to particularly thank the IDC'16 programme committee members for their hard work and dedication. To conclude, we would like to acknowledge the financial support received from Spanish Ministry of Economy and Competitiveness (MINECO) projects: EphemeCH (TIN2014-56494-C4-{1…4}-P) and DeepBio (TIN2017-85727-C4-{1…4}-P), both under the European Regional Development Fund FEDER and the support by COMPETE: POCI-01-0145-FEDER-007043 and FCT Fundao para a Cincia e Tecnologia within the Project Scope: UID/CEC/00319/2013.

Abstract Nowadays, the world is getting increasingly competitive and the quality and the amount of the work presented are one of the decisive factors when choosing an employee. It is no longer necessary to only perform but, to achieve a product with quality, on time, at the lowest possible cost and with the minimum resources. For this reason, the employee must have a high score of attention when performing a task, and the factors that influence attention negatively must be reduced. This is true in many different domains, from the workplace to the classroom. In this paper, we present a nonintrusive smart environment for monitoring people's attention when working in teams. The presented system provides real time information about each individual and information about the team. It can be very useful for team managers to identify potentially distracting events or individuals because when the attention of an individual is not at its best when performing the proposed task, her/his performance will be negatively affected, with consequences for the individual and for the organization.

S.670-675

The ability to recognize emotions in spoken words is central in human communication and social relationships. When studying one's ability to perceive emotions, the standard paradigm is to have listeners choose which one of several emotion words best characterizes linguistically neutral utterances made by actors attempting to portray various emotional states. Usually, generic experiment control software are used, which may present several limitations. In this paper we present a novel approach to the problem, based on a mobile application that can be easily configured by the researcher to set up the desired protocol. This approach not only facilitates and improves study design and data collection, but also provides a plethora of new variables about the participants that, to the best of our knowledge, have never been considered before in this domain, including behavioural research.

Abstract It is a common affair to settle disputes out of courts nowadays, through negotiation, mediation or any other mean. This has also been implemented over telecommunication means under the so‐called Online Dispute Resolution methods. However, this new technology‐supported approach is impersonal and cold, leaving aside important issues such as the disputants' body language, stress level or emotional response while being based on forms, e‐mails or chat rooms. To overcome this shortcoming, in this paper, it is proposed the creation of intelligent environments for conflict resolution that can complement the existing tools with important knowledge about the context of interaction. This will allow decision‐makers to take better framed decisions based not only on figures but also on important contextual information, similar to what happens when parties communicate in the physical presence of each other.

Abstract Ambient Assisted Living (AAL) and Ambient Intelligence (AmI) try to achieve a future where technology surrounds the users and helps them in their daily lives. In this sense, the urgent need of solutions to cover the rapid increase of the elderly population with chronic diseases led to the increase of projects related with AAL and AmI. During the latest years, several projects have been proposed to tackle different medical problems, some building devices and others services. This paper presents iGenda and its evolution, the UserAccess, with the main objective of developing an AAL platform. It features an analysis of the latest developments and points future directions for the work. These projects display the importance of the interoperability of the platforms, demonstrating a case study for AAL development.

Monitoring and managing performance in the workplace is nowadays an important aspect, in a time in which methodologies like Agile push individual and team limits further. Current performance monitoring approaches are either intrusive or based on productivity measures and are thus often dreaded by workers. Moreover, these approaches do not take into account the importance and role of the numerous external factors that influence productivity. We present a non-intrusive performance monitoring environment based on behavioral biometrics and real time analytics. It monitors and analyzes 15 features extracted from the workers’ interaction with the computer and can provide a measure of performance that is completely transparent. This measure is sensitive to external factors such as mental fatigue, stress or emotional valence. We validate this environment by assessing the effects of musical selection on Human–Computer Interaction. Results show a significant improvement on mouse motion when participants listen to the selected auditory stimuli and a negative effect on typing performance, especially with stimuli with positive tension. This work will enable the development of performance monitoring and management environments, with benefits for both organizations and individuals.

Nowadays, there is a need to provide new applications which allow the definition and implementation of safe environments that attends to the user needs and increases their wellbeing. In this sense, this paper introduces the EJaCalIVE framework which allows the creation of emotional virtual environments that incorporate agents, eHealth related devices, human actors, and emotions projecting them virtually and managing the interaction between all the elements. In this way, the proposed framework allows the design and programming of intelligent virtual environments, as well as the simulation and detection of human emotions which can be used for the improvement of the decision-making processes of the developed entities. The paper also shows a case study that enforces the need of this framework in common environments like nursing homes or assisted living facilities. Concretely, the case study proposes the simulation of a residence for the elderly. The main goal is to have an emotion-based simulation to train an assistance robot avoiding the complexity involved in working with the real elders. The main advantage of the proposed framework is to provide a safe environment, that is, an environment where users are able to interact safely with the system.

Summary Monitoring an individual's performance in a task, especially in the workplace context, is becoming an increasingly interesting and controversial topic in a time in which workers are expected to produce more, better and faster. The tension caused by this competitiveness, together with the pressure of monitoring, may not work in favour of the organization's objectives. In this paper, we present an innovative approach on the problem of performance management. We build on the fact that computers are nowadays used as major work tools in many workplaces to devise a non‐invasive method for distributed performance monitoring based on the observation of the worker's interaction with the computer. We then look at musical selection both as a pleasant and as an effective method for improving performance in the workplace. The proposed approach will allow team coordinators to assess and manage their co‐workers' performance continuously and in real‐time, using a distributed service‐based architecture. Copyright © 2015 John Wiley & Sons, Ltd.

Supporting group decision-making in ubiquitous contexts is a complex task that must deal with a large amount of factors to succeed. Here we propose an approach for an intelligent negotiation model to support the group decision-making process specifically designed for ubiquitous contexts. Our approach can be used by researchers that intend to include arguments, complex algorithms, and agents’ modeling in a negotiation model. It uses a social networking logic due to the type of communication employed by the agents and it intends to support the ubiquitous group decision-making process in a similar way to the real process, which simultaneously preserves the amount and quality of intelligence generated in face-to-face meetings. We propose a new look into this problem by considering and defining strategies to deal with important points such as the type of attributes in the multicriterion problems, agents’ reasoning, and intelligent dialogues.

In recent years, mobile devices and applications have known a growth that is unprecedented in any other technological field, reaching virtually all aspects of our lives including sports, leisure, social relationships or health. This paper describes the development of an environment to assess auditory emotional recognition based on a mobile application. The primary aim of this work is to provide a valuable instrument that can be used both in research and clinical settings, responding to the strong need of validated measures of emotional processing in Portugal. The secondary aim is to study behavioral features, acquired unobtrusively from the interaction of the participant with the device, in search for a relationship with medical conditions, cognitive impairments, auditory emotional recognition or socio-demographic indicators. This will establish the foundation for the prediction of such aspects based on the analysis of people's interaction with technological devices, providing new potentially interesting diagnostic tools.

The objectives of this work are (1) to develop a classifier application for tumor staging based on a formal representation of the Tumor-Node-Metastasis classification system (TNM), and (2) to show the feasibility of this approach on real data. This paper presents a classifier application for colorectal tumors based on the TNM-O ontology. It was developed in the JAVA using the OWL-API. The TNM-O uses the Foundational Model of Anatomy for representing anatomical entities and BioTopLite2 as a domain-top-level ontology. The classifier application processes input data via a user interface or tabular data. The classification starts with the creation of RDF Individuals for each pathological information item formally described in the ontology. These Individuals are then classified by the HermiT Description Logics reasoner by A-Box classification. A dataset with 382 entries was provided by the pathology department of a university hospital. It was automatically classified with regard to metastatic regional lymph nodes. Results or expert classification by pathologists and automatic classification were compared. The automatic process helped to detect and explain inconsistencies between expert and automatic classifications. This work, we demonstrate the use of semantic technologies in a TNM classifier application separating underlying medical knowledge represented in OWL from process logics. The presented prototypical TNM classifier application shows the potential to be integrated in larger software systems.

The role of assistive technologies is to help users with diminished capabilities in the fulfillment of their everyday tasks. One of such tasks is orientation. It is crucial for the autonomy of an individual and, at the same time, it is one of the most challenging tasks for an individual with cognitive disabilities. Existing solutions that tackle this problem are mostly concerned with guidance, tracking and the display of information. However, there is a dimension that has not been the object of concern in existing projects, the prediction of user actions. This work presents a Speculative Module for an orientation system that is used to alert the user for potential mistakes in his path, anticipating possible shifts in the wrong direction in critical points of the route. With this module, it becomes possible to issue warnings to the user and increase his attention so as to avoid a deviation from the correct path.

The availability data sources is increasing thanks to the rise of interconnected devices with sensory capabilities. Their use, has also become predominant in tasks that are not directly related with the devices' primary function or for which they may not even be designed for. Taking in consideration the act of driving, this research shows how through a composition of theories and services with these connected devices, it is possible to monitor and profile drivers performance and the generate community knowledge that can be used for the creation of user notification and warning systems. Thanks to the composition of theories from ubiquitous sensorization, machine learning, and communications it possible to design a general platform based on ubiquitous devices that generate information and models about each user behaviour individually and in communities around specific locations. This helps the improvement of not only performance but also security. Sensorization of large communities of users enrolled in driving activities can generate useful knowledge to the community. This article explores ubiquitous sensorization over the PHESS Driving platform to create reliable information that can be shared between users based on communication opportunity and location.

Clinical Practice Guidelines, as evidence-based recommendations are the ideal support for Clinical Decision Support Systems. The intricacies of a guideline execution tool are related with the establishment of a care flow with an appropriate order between procedures and the modelling of decision points. One of such decision points is the choice between alternative tasks based on trigger conditions regarding a patient's state. It may be the case that, when there is the need to choose one of the alternative tasks, the system does not possess all the required information to do so, thus rendering impossible to reach an outcome. Speculative Computation and Abduction may increase the efficiency of this process by allowing the system to advance the computation of a solution, even while it is waiting for a response from the information sources. This work provides the basis for a Speculative Computation framework able to cope with decisions of clinical care flows. The methods developed herein were devised to support practitioners and to improve patient-centred medicine by providing maps of the most likely evolution of a patient, even when the information is incomplete.

Recent research shows that our performance and satisfaction at work depends more on motivational factors than the number of hours or the intensity of the work. In this paper we propose a framework aimed at managing motivation to improve workplace indicators. The key idea is to allow team managers and workers to negotiate over the conditions of the tasks so as to find the best motivation for the worker within the constraints of what the organization may offer.

Negotiation is a collaborative activity that requires the participation of different parties whose behaviors influence the outcome of the whole process. The work presented here focuses on the identification of such behaviors and their impact on the negotiation process. The premise for this study is that identifying and cataloging the behavior of parties during a negotiation may help to clarify the role that stress plays in the process. To do so, an experiment based on a negotiation game was implemented. During this experiment, behavioral and contextual information about participants was acquired. The data from this negotiation game were analyzed in order to identify the conflict styles used by each party and to extract behavioral patterns from the interactions, useful for the development of plans and suggestions for the associated participants. The work highlights the importance of the knowledge about social interactions as a basis for informed decision support in situations of conflict.

The level of uncertainty and incompleteness in the information upon which healthcare professionals have to make judgments has been a subject of discussion in the past, and more nowadays, with the advent of the so-called Clinical Decision Support Systems. This work addresses uncertainty in the postoperative prognosis for colorectal cancer. The interdependence and synergistic effect of different clinical features comes into play when it is necessary to predict how a patient will react to this type of surgery. Using a probabilistic based knowledge representation, a decision support system was conceived in order to provide support for physicians under these circumstances, in particular to surgeons. The solution proposed is based on machine learning on records of cancer patients, incorporating explicit knowledge of experts about the domain. To facilitate access and thus increase its dissemination in the healthcare community, the system is integrated in a wider platform available through a web application.

The exponential increase of home-bound persons thatlive alone and are in need of continuous monitoring requires newsolutions to current problems. Most of these cases presentillnesses, such as motor or psychological disabilities, that deprivethem of a normal living. Abnormal situations such asforgetfulness or falls are quite common and should be preventedor dealt with. This paper presents a system able to detectdangerous situations at home, such as falls, independently fromexisting environment conditions. The aim of the proposed systemis to proactively offer support to the citizen or to warn theemergency services when needed.

This article proposes a process for automatic population of ontologies from text that applies natural language processing and information extraction techniques to acquire and classify ontology instances. The work is part of HERMES, an FCT/CAPES research project looking for techniques and tools for automating the process of ontology learning and population. Two experiments using a legal and a tourism corpora were conducted in order to evaluate it. The results indicate that our approach can extract and classify instances with high effectiveness with the additional advantage of domain independence.

Information Retrieval is a theme that is so multifaceted as it is its significance to any knowledge-based sphere of influence. This is true in The Law, especially when we judge under the angle of the so-called On-line Dispute Resolution. Indeed, there is the need to analyze and develop efficient information retrieval methods that may improve the course of actions that depend on such techniques. It was under this line of thought that we look at two different methods for information retrieval, and then strengthen its advantages into a third one. The results of this effort are now being applied in UMCourt, an Online Dispute Resolution platform that helps disputant parties and software agents to interact and make their decisions.

The use of technology to support conflict resolution is nowadays well established. Moreover, technological solutions are not only used to solve traditional conflicts but also to solve conflicts that emerge in virtual environments. Therefore, a new field of research has been developing in which the use of Artificial Intelligence techniques can significantly improve the conflict resolution process. In this paper we focus on developing conflict resolution models that are able to classify the disputant parties according to their personal conflict style. Moreover, we present a dynamic conflict resolution model that is able to use that information to adapt strategies in real time according to significant changes in the context of interaction. To do it we follow a novel approach in which an intelligent environment supports the lifecycle of the conflict resolution model with the provision of important context knowledge.

The advances observed in the last years in telecommunication technologies rapidly brought along new ways of doing business. This new reality, however, has not been so rapidly followed by the entities responsible for dealing with the conflicts that arise from these interactions, now undertaken in an electronic format. Traditional paper-based courts, designed for the industrial era, are now outdated. The answer to this problem may rely on the new tools that can be built using new artifacts from fields such as Artificial Intelligence. Using these tools the parties can simulate outcomes, thus having a better notion of the possible consequences of a legal dispute, namely in terms of the Best and Worst Alternative to Negotiated Agreements. In this paper, we present our agent-based architecture for such a tool, UMCourt, placing special emphasis on a particular agent that, based on the concept of legal precedent, gives its users a set of possible outcomes of a case, based on the observation of past similar cases and learns new cases in order to enrich its knowledge base about the Portuguese labor law.

According to Portuguese Law (Decree 7/2004 article 33) the general rules on error in electronic contracting without human intervention will apply only in case of human error in programming (it will be applied the rules considered for human error in the formation of the will), in case of malfunctioning of the machine (it will be applied the rules concerning error on declaration) and if the message does not arrive to the destination exactly as it was sent (it will be applied the rules on machine error on transmission).

Recent trends in the field of Artificial Intelligence, brought along new ways of formalizing and expressing wills and declarations. Its application to Virtual Enterprises requires an analysis of the interactions among agents, frameworks and users, as well as technical and legal analysis, in order to discover the rules to be applied, to solve a particular problem under a prospective scenario. Credibility, trust and security issues must be taken under consideration, especially concerning authenticity, confidentiality, integrity and non-repudiation. In order to increase the use of agents in Virtual Enterprises, besides the analysis and research of legal solutions in the commercial arena, it is essential to assure that agents will meet requirements of credibility and trust, insuring a transparent and secure way for their commercial acting, now capable of generating legal relations. This paper shows how to construct a dynamic virtual world of complex and interacting entities or agents, in which fitness is judged by a quality of information criterion. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.