Abstract Tools for automatic wood species identification are needed worldwide in order to support sustainable timber trade. This work explores the application of computer vision techniques to classify high-resolution macroscopic images of timber. The main challenge of this problem is that fine-grained patterns in timber are crucial in order to accurately identify wood species, and these patterns are not learned by convolutional neural networks (CNNs) trained on low resolution images. This work introduces the Timber Deep Learning Identification with Patch-based Inference Voting methodology, abbreviated TDLI-PIV methodology. This methodology exploits the concept of patching and the availability of high-resolution macroscopic images of timber in order to overcome the inherent challenges that CNNs face in timber identification. The TDLI-PIV methodology is able to capture fine-grained patterns in timber and, moreover, boosts robustness and prediction accuracy via a collaborative voting inference process. In this work we also introduce a new data set of marcroscopic images of timber, called GOIMAI-Phase-I, which has been obtained using optical magnification, and is openly published online in zenodo. Our experiments have assessed the performance of the TDLI-PIV methodology, including a comparison with other methodologies available in the literature, an exploration of data augmentation methods and the effect that the dataset size has on the accuracy of TDLI-PIV.

In cross-silo Federated Learning (FL), client selection is critical to ensure high model performance, yet it remains challenging due to data quality decompensation, budget constraints, and incentive compatibility. As training progresses, these factors exacerbate client heterogeneity and degrade global performance. Most existing approaches treat these challenges in isolation, making it difficult to optimize multiple factors in conjunction. To address this, we propose Shapley-Bid Reputation Optimized Federated Learning (SBRO-FL), a unified framework integrating dynamic bidding, reputation modeling, and cost-aware selection. Clients submit bids based on their perceived data quality, and their contributions are evaluated using Shapley values to quantify their marginal impact on the global model. A reputation system, inspired by prospect theory, captures historical performance while penalizing inconsistency. The client selection problem is formulated as a 0-1 integer program that maximizes reputation-weighted utility under budget constraints. Experiments on four benchmark datasets demonstrate the framework’s effectiveness, improving final model accuracy by an average of 10.3% over random selection, with gains exceeding 19% on more complex datasets like CIFAR-10 and SVHN. Our results highlight the importance of balancing data reliability, incentive compatibility, and cost efficiency to enable scalable and trustworthy FL deployments.

The study of large language models (LLMs) is a key area in open-world machine learning. Although LLMs demonstrate remarkable natural language processing capabilities, they also face several challenges, including consistency issues, hallucinations, and jailbreak vulnerabilities. Jailbreaking refers to the crafting of prompts that bypass alignment safeguards, leading to unsafe outputs that compromise the integrity of LLMs. This work specifically focuses on the challenge of jailbreak vulnerabilities and introduces a novel taxonomy of jailbreak attacks grounded in the training domains of LLMs. It characterizes alignment failures as arising from gaps in generalization, objectives, and robustness. Our primary contribution is a perspective on jailbreak, framed through the different linguistic domains that emerge during LLM training and alignment. This viewpoint highlights the limitations of existing approaches and enables us to classify jailbreak attacks in terms of the underlying model deficiencies they exploit. Unlike conventional classifications that categorize attacks based on prompt construction methods (e.g., prompt templating), our approach provides a deeper understanding of LLM behavior. We introduce a taxonomy with four categories—mismatched generalization, competing objectives, adversarial robustness, and mixed attacks— offering insights into the fundamental nature of jailbreak vulnerabilities. Finally, we present key lessons derived from this taxonomic study. • We perform an analysis on why aligned Large Language Models (LLMs) are vulnerable to jailbreaking attacks. These attacks allows an user to generate answers against the policies of a LLM company. The analysis is done from a domain perspective, where we distinguish different training domain regions. This analysis is based and extended from the Jailbroken hypothesis, a paper published at the NeurIPS conference. • Based on the previous analysis, we propose a taxonomy to classify LLMs jailbreaking attacks. Three main types of attacks are distinguished, namely mismatched generalization, competing objectives and adversarial robustness. We further categorize these types of attacks into subgroups. • We extensively review the LLM jailbreaking literature to support our taxonomy, classifying them into each of the categories we propose.

Out-of-distribution (OOD) detection is a critical task in machine learning, particularly in safety-sensitive applications where model failures can have serious consequences. However, current OOD detection methods often suffer from restrictive distributional assumptions, limited scalability, and a lack of interpretability. To address these challenges, we propose STOOD-X , a two-stage methodology that combines a Statistical nonparametric Test for OOD Detection with eXplainability enhancements. In the first stage, STOOD-X uses feature-space distances and a Wilcoxon-Mann-Whitney test to identify OOD samples without assuming a specific feature distribution. In the second stage, it generates user-friendly, concept-based visual explanations that reveal the features driving each decision, aligning with the BLUE XAI paradigm. Through extensive experiments on benchmark datasets and multiple architectures, STOOD-X achieves competitive performance compared to state-of-the-art post hoc OOD detectors, particularly in high-dimensional and complex settings. In addition, its explainability framework enables human oversight, bias detection, and model debugging, fostering trust and collaboration between humans and AI systems. Therefore, STOOD-X offers a robust, explainable, and scalable solution for real-world OOD detection tasks.

In Artificial Intelligence, there is an increasing demand for adaptive models capable of dealing with a diverse spectrum of learning tasks, surpassing the limitations of systems devised to cope with a single task. The recent emergence of General-Purpose Artificial Intelligence Systems (GPAIS) poses model configuration and adaptability challenges at far greater complexity scales than the optimal design of traditional Machine Learning models. Evolutionary Computation (EC) has been a useful tool for both the design and optimization of Machine Learning models, endowing them with the capability to configure and/or adapt themselves to the task under consideration. Therefore, their application to GPAIS is a natural choice. This paper aims to analyze the role of EC in the field of GPAIS, exploring the use of EC for their design or enrichment. We also match GPAIS properties to Machine Learning areas in which EC has had a notable contribution, highlighting recent milestones of EC for GPAIS. Furthermore, we discuss the challenges of harnessing the benefits of EC for GPAIS, presenting different strategies to both design and improve GPAIS with EC, covering tangential areas, identifying research niches, and outlining potential research directions for EC and GPAIS.

Evolutionary computation has been successfully applied to tackle a wide variety of machine learning problems due to its generalisation and adaptability capabilities. Recently, it has shown great potential to enhance General Purpose Artificial Intelligence Systems particularly, those that work in open-world scenarios which require dynamic adaptation abilities. Zero-Shot Learning (ZSL) is an emerging paradigm within the open-world context that allows us to perform predictive tasks, such as the classification of unknown elements (i.e. unknown classes) for which a model has not been specifically trained. To do this, ZSL uses auxiliary information known as semantic space, typically in the form of attributes that define each class. This plays a crucial role in associating prior knowledge with unknown situations. However, the treatment of the semantic space has remained an underexplored area, as selecting the most relevant semantic attributes that generalise to unknown classes is a challenging problem. In this preliminary work, we propose a tailored genetic algorithm to perform feature selection of the semantic space, removing irrelevant features that negatively affect the generalisation capabilities of a well-known ZSL approach. The results on four commonly used ZSL image classification problems show that refining the semantic space may consistently boost the accuracy across all datasets.

The decentralized nature of federated learning (FL) poses critical challenges related to security: Clients participating in the process may not necessarily be trustworthy and could engage in adversarial attacks, potentially undermining the integrity and reliability of the global machine learning model. Security concerns have been extensively investigated in traditional FL, where collaboratively learned models are typically deep neural networks. However, this class of models does not meet the requirement of explainability, which is considered essential for the trustworthiness of AI systems. In this work, we present an analysis on security threats to FL of explainable models, namely fuzzy rule-based classifiers (FRBCs). We outline the types of attacks a malicious client may implement, and assess, through a preliminary experimental analysis, the impact they have on FL of FRBCs in terms of global model performance. We also compare these findings with the effects of the same or similar well-established attacks in traditional FL of neural network models. Finally, we provide insights to improve the security of FRBCs learned in a federated fashion.

With the rising usage of artificial intelligence systems, social concerns around them and the need for regulation is also increasing, including requirements for data privacy. Federated Learning addresses data privacy concerns in distributed machine learning by training models collaboratively without centralizing data. However, Federated Learning is susceptible to Byzantine attacks, where malicious nodes submit corrupted updates. Krum, a robust aggregation algorithm, has been widely adopted as a defense mechanism. However, recent studies have shown that Krum’s performance degrades significantly in high-dimensional settings. This work proposes Layerwise Krum, a novel aggregation method that enhances Krum’s robustness in high-dimensional spaces while maintaining computational efficiency. We provide theoretical analysis of the improved robustness of Layerwise Krum compared to standard Krum. Furthermore, we empirically evaluate Layerwise Krum on various image classification datasets under diverse data distributions and Byzantine attack scenarios, consistently demonstrating superior performance compared to the original Krum operator.

Artificial intelligence (AI) systems are increasingly adopted in high‐stakes domains such as healthcare and finance, so the demand for transparency and interpretability has grown substantially. EXplainable AI (XAI) methods have emerged to address this challenge, but individual techniques often offer limited, fragmented insights. This paper introduces Meta‐explainers, a novel ensemble‐based XAI framework that integrates multiple explanation types—specifically relevance‐based and counterfactual methods—into unified, multifaceted and complementary meta‐explanations. Inspired by meta‐classification principles, our approach structures the explanation process into five stages: generation, grouping, evaluation, aggregation, and visualization. Each stage is designed to preserve the unique strengths of individual XAI techniques while enhancing their interpretability and coherence when combined. Experimental results on both image (MNIST) and tabular (Breast Cancer) datasets show that Meta‐explainers consistently outperform individual and state‐of‐the‐art ensemble explanation methods in terms of explanation quality, as measured by established metrics. This work paves the way toward more holistic and user‐centered AI explainability with a flexible methodology that can be extended to incorporate additional explanation paradigms.

The increasing adoption of artificial intelligence (AI) in critical domains such as healthcare, law, and defense demands robust mechanisms to ensure transparency and explainability in decision‐making processes. While machine learning and deep learning algorithms have advanced significantly, their growing complexity presents persistent interpretability challenges. Existing maturity frameworks, such as Capability Maturity Model Integration, fall short in addressing the distinct requirements of explainability in AI systems, particularly where ethical compliance and public trust are paramount. To address this gap, we propose the Maturity Model for eXplainable Artificial Intelligence: Analysis and Evaluation (MM4XAI‐AE), a domain‐agnostic maturity model tailored to assess and guide the practical deployment of explainability in AI‐based applications. The model integrates two complementary components: an analysis model and an evaluation model, structured across four maturity levels—operational, justified, formalized, and managed. It evaluates explainability across three critical dimensions: technical foundations, structured design, and human‐centered explainability. MM4XAI‐AE is grounded in the PAG‐XAI framework, emphasizing the interrelated dimensions of practicality, auditability, and governance, thereby aligning with current reflections on responsible and trustworthy AI. The MM4XAI‐AE model is empirically validated through a structured evaluation of thirteen published AI applications from diverse sectors, analyzing their design and deployment practices. The results show a wide distribution across maturity levels, underscoring the model’s capacity to identify strengths, gaps, and actionable pathways for improving explainability. This work offers a structured and scalable framework to standardize explainability practices and supports researchers, developers, and policymakers in fostering more transparent, ethical, and trustworthy AI systems.

Anomaly detection is the process of identifying observations that differ greatly from the majority of data. Unsupervised anomaly detection aims to find outliers in data that is not labeled, therefore, the anomalous instances are unknown. The exponential data generation has led to the era of Big Data. This scenario brings new challenges to classic anomaly detection problems due to the massive and unsupervised accumulation of data. Traditional methods are not able to cop up with computing and time requirements of Big Data problems. In this paper, we propose four distributed algorithm designs for Big Data anomaly detection problems: HBOS_BD, LODA_BD, LSCP_BD, and XGBOD_BD. They have been designed following the MapReduce distributed methodology in order to be capable of handling Big Data problems. These algorithms have been integrated into an Spark Package, focused on static and dynamic Big Data anomaly detection tasks, namely AnomalyDSD. Experiments using a real-world case of study have shown the performance and validity of the proposals for Big Data problems. With this proposal, we have enabled the practitioner to efficiently and effectively detect anomalies in Big Data datasets, where the early detection of an anomaly can lead to a proper and timely decision.

The rapid development of artificial intelligence systems has amplified societal concerns regarding their usage, necessitating regulatory frameworks that encompass data privacy. Federated learning (FL) is proposed as a potential solution to the challenges of data privacy in distributed machine learning by enabling collaborative model training without data sharing. However, FL systems remain vulnerable to Byzantine attacks, where malicious nodes contribute corrupted model updates. Although Byzantine resilient rules have emerged as a widely adopted robust aggregation algorithm to mitigate these attacks, their effectiveness drops significantly in high-dimensional parameter spaces, sometimes leading to poor-performing models. This paper introduces Layerwise Cosine Aggregation, a novel aggregation scheme designed to enhance the robustness of these rules in such high-dimensional settings while preserving computational efficiency. A theoretical analysis is presented, demonstrating the superior robustness of the proposed Layerwise Cosine Aggregation compared to the original robust aggregation rules. Empirical evaluation in diverse image classification datasets, under varying data distributions and Byzantine attack scenarios, consistently demonstrates the improved performance of Layerwise Cosine Aggregation, achieving up to a 16% increase in model accuracy.

The emergence of deep learning at the beginning of the last decade has driven the advancement of complex models, culminating in the development of large language models and generative AI.These models represent the summit of size and complexity.Explainability should be an option that plays a key role in enabling understandable the AI-assisted decision-making and ensuring accountability.This contribution delves into the complexities of explainable artificial intelligence (XAI) through various perspectives, considering the extensive and growing body of literature.Our discussion begins by addressing the challenges posed by complex data, models, and high-risk scenarios.Given the rapid growth of the field, it is essential to tackle the criticisms and challenges that emerge as it matures, requiring thorough exploration.This contribution explores them, along with three aspects that may shed light on them.First, it is focused on the lack of definitional cohesion, examining how and why is defined XAI from the perspectives of audience and understanding.Second, it explores XAI explanations, bridging the gap between complex AI models and human understanding.Third, it is crucial to consider how to analyze and evaluate the maturity level of explainability, from a triple dimension, practicality, governance and auditability. Index Terms-eXplainable

Transformers have emerged as a highly effective architecture for natural language processing and computer vision. Of late, there has been a surge in initiatives aimed at refining this architecture to enhance its applicability to long sequence time-series forecasting, yielding promising outcomes.This paper introduces Local Attention, an efficient attention mechanism tailored for time series data. This mechanism exploits the continuity properties of time series and the principle of locality in order to compute less attention scores. We provide an Θ(n log n) algorithm to implement Local Attention based on tensor algebra results, which contrasts to the Θ(n2) time and memory complexity of the original attention mechanism.Our experimental analysis shows that the vanilla transformer with Local Attention outperforms state of the art models based on probabilistic attention mechanisms. These findings affirm the effectiveness of our approach and outline a spectrum of future challenges in long sequence time series forecasting.

Significant advancements in the field of wood species recognition are needed worldwide to support sustainable timber trade. In this work we contribute to automate the identification of wood species using machine learning techniques and high-resolution macroscopic images of timber. The main challenge of this problem is that fine-grained patterns in timber are crucial in order to accurately identify wood species, and these patterns are not properly learned by traditional convolutional neural networks (CNNs) trained on low/medium resolution images.To this end, in the context of funded project GoIMAI, we have collected a Wood Species dataset to cover 37 wood species. Most of them CITES-listed and some similar species of the same genera. We propose a patch-based classification with a voting mechanism that overcomes the inherent challenges that CNNs face in timber identification. Our method is able to capture finegrained patterns in timber and boost robustness and prediction accuracy of the model. Our experiments have assessed the performance of the method, involving the comparison of several model architectures proposed in the literature, the analysis of the parameters of our proposed methodology, and the exploration of data augmentation methods.

When deploying machine learning models on resource-constrained hardware, reducing the memory footprint required by the model without compromising its performance is critical. Moreover, in open-world scenarios models often operate in dynamic and unpredictable environments where the data distribution evolves over time. Robust models can generalize well to unforeseen circumstances, including out-of-distribution inputs that may not have been encountered during the training phase. This adaptability is essential to handle the inherent variability of real-world data. This work formulates a multi-objective optimization problem that aims at optimizing the quantization resolution of the parameters of an already trained machine learning model based on three conflicting goals: maximizing the performance of the model on its designated learning task, minimizing the memory footprint of the compressed model, and enhancing its robustness against out-of-distribution data. Given the complexity of the resulting combinatorial optimization problem, we employ multi-objective evolutionary algorithms to efficiently obtain an approximation of the Pareto front balancing among the aforementioned objectives. Experiments with a randomized neural network compressed under the proposed formulation are run over several benchmark classification datasets. Different multi-objective solvers are employed to compare their effectiveness in terms of the convergence and diversity of their produced Pareto estimations. Additionally, we assess the achieved equilibrium between the three objectives against a floating-point implementation of the same model. Our experiments reveal that both the computational resources and the robustness of the model can be optimized via evolutionary quantization without significantly sacrificing its performance for the task at hand.

Abstract This paper presents, for the first time, a distance metric learning algorithm for monotonic classification. Monotonic datasets arise in many real-world applications, where there exist order relations in the input and output variables, and the outputs corresponding to ordered pairs of inputs are also expected to be ordered. Monotonic classification can be addressed through several distance-based classifiers that are able to respect the monotonicity constraints of the data. The performance of distance-based classifiers can be improved with the use of distance metric learning algorithms, which are able to find the distances that best represent the similarities among each pair of data samples. However, learning a distance for monotonic data has an additional drawback: the learned distance may negatively impact the monotonic constraints of the data. In our work, we propose a new model for learning distances that does not corrupt these constraints. This methodology will also be useful in identifying and discarding non-monotonic pairs of samples that may be present in the data due to noise. The experimental analysis conducted, supported by a Bayesian statistical testing, demonstrates that the distances obtained by the proposed method can enhance the performance of several distance-based classifiers in monotonic problems.

Deep learning models like chatGPT exemplify AI success but necessitate a deeper understanding of trust in critical sectors. Trust can be achieved using counterfactual explanations, which is how humans become familiar with unknown processes; by understanding the hypothetical input circumstances under which the output changes. We argue that the generation of counterfactual explanations requires several aspects of the generated counterfactual instances, not just their counterfactual ability. We present a framework for generating counterfactual explanations that formulate its goal as a multiobjective optimization problem balancing three objectives: plausibility; the intensity of changes; and adversarial power. We use a generative adversarial network to model the distribution of the input, along with a multiobjective counterfactual discovery solver balancing these objectives. We demonstrate the usefulness of six classification tasks with image and 3D data confirming with evidence the existence of a trade-off between the objectives, the consistency of the produced counterfactual explanations with human knowledge, and the capability of the framework to unveil the existence of concept-based biases and misrepresented attributes in the input domain of the audited model. Our pioneering effort shall inspire further work on the generation of plausible counterfactual explanations in real-world scenarios where attribute-/concept-based annotations are available for the domain under analysis.

Abstract The annotation of ambiguous or subjective NLP tasks is usually addressed by various annotators. In most datasets, these annotations are aggregated into a single ground truth. However, this omits divergent opinions of annotators, hence missing individual perspectives. We propose FLEAD (Federated Learning for Exploiting Annotators’ Disagreements), a methodology built upon federated learning to independently learn from the opinions of all the annotators, thereby leveraging all their underlying information without relying on a single ground truth. We conduct an extensive experimental study and analysis in diverse text classification tasks to show the contribution of our approach with respect to mainstream approaches based on majority voting and other recent methodologies that also learn from annotator disagreements.

Airway segmentation is crucial for the examination, diagnosis, and prognosis of lung diseases, while its manual delineation is unduly burdensome. To alleviate this time-consuming and potentially subjective manual procedure, researchers have proposed methods to automatically segment airways from computerized tomography (CT) images. However, some small-sized airway branches (e.g., bronchus and terminal bronchioles) significantly aggravate the difficulty of automatic segmentation by machine learning models. In particular, the variance of voxel values and the severe data imbalance in airway branches make the computational module prone to discontinuous and false-negative predictions, especially for cohorts with different lung diseases. The attention mechanism has shown the capacity to segment complex structures, while fuzzy logic can reduce the uncertainty in feature representations. Therefore, the integration of deep attention networks and fuzzy theory, given by the fuzzy attention layer, should be an escalated solution for better generalization and robustness. This article presents an efficient method for airway segmentation, comprising a novel fuzzy attention neural network (FANN) and a comprehensive loss function to enhance the spatial continuity of airway segmentation. The deep fuzzy set is formulated by a set of voxels in the feature map and a learnable Gaussian membership function. Different from the existing attention mechanism, the proposed channel-specific fuzzy attention addresses the issue of heterogeneous features in different channels. Furthermore, a novel evaluation metric is proposed to assess both the continuity and completeness of airway structures. The efficiency, generalization, and robustness of the proposed method have been proved by training on normal lung disease while testing on datasets of lung cancer, COVID-19, and pulmonary fibrosis.

Consistency is usually associated with transitive properties, among which the additive transitivity is one of the most popular methods. Although various linguistic additive consistency measurements have been proposed in the literature, the existing additive consistency measurements are usually directly proposed without a general definition with axiomatic design, and thus there is a lack of axiomatic comparisons among them. Moreover, the existing additive consistency measurements seldom consider the general case of personalized individual semantics (PIS) where a word has different numerical meanings to different people. In this study, we aim to develop a comparative study for the linguistic additive consistency measurements in the PIS context from an axiomatic perspective. Specifically, we first revisit the most widely used additive consistency measurements of simple linguistic preference relations (SLPRs) and hesitant fuzzy linguistic preference relations (HFLPRs) in the PIS context. Next, we present several rational axioms that the additive consistency measurements with PIS in SLPRs and HFLPRs should satisfy. Finally, we compare these additive consistency measurements by verifying the satisfaction of the proposed axioms. The results in this article show that the existing linguistic additive consistency measurements behave differently from the axiomatic perspective: they perform well in SLPRs but cannot satisfy all axioms in HFLPRs.

Explainable artificial intelligence is proposed to provide explanations for reasoning performed by artificial intelligence. There is no consensus on how to evaluate the quality of these explanations, since even the definition of explanation itself is not clear in the literature. In particular, for the widely known local linear explanations, there are qualitative proposals for the evaluation of explanations, although they suffer from theoretical inconsistencies. The case of image is even more problematic, where a visual explanation seems to explain a decision while detecting edges is what it really does. There are a large number of metrics in the literature specialized in quantitatively measuring different qualitative aspects, so we should be able to develop metrics capable of measuring in a robust and correct way the desirable aspects of the explanations. Some previous papers have attempted to develop new measures for this purpose. However, these measures suffer from lack of objectivity or lack of mathematical consistency, such as saturation or lack of smoothness. In this paper, we propose a procedure called REVEL to evaluate different aspects concerning the quality of explanations with a theoretically coherent development which do not have the problems of the previous measures. This procedure has several advances in the state of the art: it standardizes the concepts of explanation and develops a series of metrics not only to be able to compare between them but also to obtain absolute information regarding the explanation itself. The experiments have been carried out on four image datasets as benchmark where we show REVEL’s descriptive and analytical power.

The combination of convolutional and recurrent neural networks is a promising framework. This arrangement allows the extraction of high-quality spatio-temporal features together with their temporal dependencies. This fact is key for time series prediction problems such as forecasting, classification or anomaly detection, amongst others. In this paper, the TSFEDL library is introduced. It compiles 22 state-of-the-art methods for both time series feature extraction and prediction, employing convolutional and recurrent deep neural networks for its use in several data mining tasks. The library is built upon a set of Tensorflow + Keras and PyTorch modules under the AGPLv3 license. The performance validation of the architectures included in this proposal confirms the usefulness of this Python package.

Anomaly detection is the task of detecting samples that behave differently from the rest of the data or that include abnormal values. Unsupervised anomaly detection is the most common scenario, which implies that the algorithms cannot train with a labeled input and do not know the anomaly behavior beforehand. Histogram-based methods are one of the most approaches in unsupervised anomaly detection, remarking a good performance and a low runtime. Despite the good performance, histogram-based anomaly detectors are not capable of processing data flows while updating their knowledge and cannot deal with a high amount of samples. In this paper, we propose a new histogram-based approach for addressing the aforementioned problems by introducing the ability to update the information inside a histogram. We have applied these strategies to design a new algorithm called Multi-step Histogram Based Outlier Scores (MHBOS), including five new histogram update mechanisms. The results have shown the performance and validity of MHBOS as well as the proposed strategies in terms of performance and computing times.

In group decision making (GDM), opinion dynamics is a useful tool to investigate consensus formation. Notably, consensus convergence speed is of key importance to manage the consensus formation in GDM with opinion dynamics. Recently, social network DeGroot (SNDG) model has been widely used in opinion dynamics. Based on this, this article dedicates to study how agents’ high self-confidence levels affect the consensus convergence speed in SNDG model. Interestingly, using theoretical analysis, we prove that: 1) the speed of consensus reaching is subject to the largest self-confidence level of opinion followers and 2) the speed of consensus reaching is also subject to the top two self-confidence levels of opinion leaders. Furthermore, through extensive simulation’, we find that the theoretical results are robust to the topological structure and the size of social networks.

Every decision may involve risks. Real-world risk issues are usually supervised by third parties. Decision-making may be affected by the absence of sufficient or reasonable trust or to the opposite, an unconditional, excessive, or blind trust, which is called trust risks. The conflict-eliminating process (CEP) aims to facilitate satisfactory consensus by decision makers (DMs) through continuous reconciliation between their opinion differences on the subject matter. This article addresses trust risks in CEP of social network group decision making (SNGDM) through third-party monitoring. A trust risk analysis-based conflict-eliminating model for SNGDM is developed. It is assumed that a third-party agency monitors the DMs' credibility and performance, which is recorded in an objective evaluation matrix and multi-attribute trust assessment matrix (MTAM). A trust risk measurement methodology is proposed to classify the DMs' different trust risk types and to measure the trust risk index (TRI) of a group of DMs. When TRI is unacceptable, a trust risk management mechanism that controls TRI is activated. Different management policies are applicable to DMs' different trust risk types. There are two main methods: 1) dynamically update the MTAM based on DMs' performance and 2) provide suggestions for modifying the DM's information with high TRI. Besides, as part of the integrated CEP, this model includes an optimization approach to dynamically derive DMs' reliable aggregation weights from their MTAM. Simulation experiments and an illustrative example support the feasibility and validity of the proposed model for managing trust risks in CEP of SNGDM.

Support vector machines (SVMs) are popular learning algorithms to deal with binary classification problems. They traditionally assume equal misclassification costs for each class; however, real-world problems may have an uneven class distribution. This article introduces EBCS-SVM: evolutionary bilevel cost-sensitive SVMs. EBCS-SVM handles imbalanced classification problems by simultaneously learning the support vectors and optimizing the SVM hyperparameters, which comprise the kernel parameter and misclassification costs. The resulting optimization problem is a bilevel problem, where the lower level determines the support vectors and the upper level the hyperparameters. This optimization problem is solved using an evolutionary algorithm (EA) at the upper level and sequential minimal optimization (SMO) at the lower level. These two methods work in a nested fashion, that is, the optimal support vectors help guide the search of the hyperparameters, and the lower level is initialized based on previous successful solutions. The proposed method is assessed using 70 datasets of imbalanced classification and compared with several state-of-the-art methods. The experimental results, supported by a Bayesian test, provided evidence of the effectiveness of EBCS-SVM when working with highly imbalanced datasets.

In realistic decision-making environments, human behaviors bring influences to various decision-making procedures, and classic multiattribute decision-making approaches based on utility decisions own some deviations from actual situations. The behavioral decision theory modifies classic decision-making theories to make the new method more applicable. Regret theory, as one of the important components in behavioral decision theories, has been widely used in theories and applications. Based on regret theory, we establish a generalized three-way decision method on incomplete multiscale decision information systems with interval fuzzy numbers. First, we select an incomplete optimal subsystem for the incomplete multiscale information system, and convert the multiscale evaluation information into an interval fuzzy number by using a linguistic term set. Second, based on the probability distribution of evaluation values and tradeoff factors, we propose a target-dependent approximation estimation method for the incomplete interval fuzzy subsystem. Then, the regret–rejoicing preferences between objects are obtained. Finally, a tripartition and ranking method based on a max-bipartition and preference index is established. Moreover, the incompleteness experiments show that the decision-making results of our method can still maintain more than 97% consistency in an incomplete information system with a missing rate of at most 20%. In addition, the parameter analysis shows that the proposed method maintains an accuracy rate of more than 98% and a classification error rate of no more than 0.6%.

Most existing opinion formation processes apply one opinion dynamics model. However, this article combines opinion formation and complex networks to innovatively develop two new opinion dynamics models to more realistically describe the opinion evolution process: 1) an opinion similarity mixed (OSM) model and 2) a structural similarity mixed (SSM) model, both of which include characteristics from the DeGroot model and the Hegselmann–Krause bounded confidence model. In addition, the strong and weak relations between individuals are considered. The network dynamically changes by two developed network updating algorithms based on opinion similarity and structural similarity. Simulations are then conducted using artificial and real-world networks, which are Erdös-Rényi random networks, random regular networks, scale-free networks, and the Twitter network. It is found that compared with static networks, the opinion evolution in dynamic networks produces fewer opinion clusters and smaller opinion variances. The dynamic network mechanism reduces the weak relations between agents and improves the global clustering coefficient in the ER random networks but not in the Twitter network, which means that the network topology has an impact on results. Therefore, it is concluded that agents’ subjective behaviors significantly influence the outcome of opinion evolution and networks, which is consistent with real life.

The “wisdom of the crowd” theory states that a nonexpert crowd makes smarter decisions than a reduced set of experts. Social network platforms are a source of evaluations in the natural language of any topic, which may be considered as the evaluations of a nonexpert crowd. Decision-making (DM) models are constrained by their inability of processing large amounts of evaluations in natural language, as those ones from social networks. We claim that evaluations from social networks can enhance the quality of multiperson multicriteria DM models. Accordingly, we propose a crowd DM model guided by sentiment analysis (SA), which solves decision situations leveraging the wisdom of the crowd available in social networks. The model uses several deep-learning SA classification models through opinion triplets to incorporate all the evaluation shades in the DM model. Likewise, the likely lack of information stemmed from the consideration of a large set of users is tackled with a sparse representation of the evaluations. We annotate and release the TripR-2020Large dataset, and we use it to evaluate the model in the use case of restaurant choice. The results show that the integration of the wisdom of the crowd and the different shades of the evaluations enhances the quality of the decision.

Modern machine learning algorithms excel in a great variety of tasks, but at the same time, it is also known that those complex models need to deal with uncertainty from different sources. Consequently, understanding if the model is indeed making accurate predictions or simply guessing at random is not trivial, and measuring the confidence bounds becomes very important. Bayesian machine learning seems to provide the solution, however, many of the state-of-the-art Bayesian algorithms use rigid parametric representations of the uncertainty where the learning process depends on the gradient of a predefined cost function. In this article, a new gradient-free training algorithm based on Approximate Bayesian Computation by Subset Simulation is proposed, where the likelihood function and the weights are defined by non-parametric formulations, resulting in a flexible and fairer representation of the uncertainty. The experiments, specially the engineering case study on composite materials subject to fatigue damage, show the ability of the proposed algorithm to consistently reach accurate predictions while avoiding gradient related instabilities, and most importantly, it provides a realistic and coherent quantification of the uncertainty represented by confidence bounds. All this may lead to a reduction of safety factors in engineering problems, and in general, allows us to make well-informed decisions in situations with a high degree of uncertainty and risk. A comparison with the state-of-the-art Bayesian Neural Networks is also carried out.

Traditionally, Convolutional Neural Networks make use of the maximum or arithmetic mean in order to reduce the features extracted by convolutional layers in a downsampling process known as pooling. However, there is no strong argument to settle upon one of the two functions and, in practice, this selection turns to be problem dependent. Further, both of these options ignore possible dependencies among the data. We believe that a combination of both of these functions, as well as of additional ones which may retain different information, can benefit the feature extraction process. In this work, we replace traditional pooling by several alternative functions. In particular, we consider linear combinations of order statistics and generalizations of the Sugeno integral, extending the latter's domain to the whole real line and setting the theoretical base for their application. We present an alternative pooling layer based on this strategy which we name "CombPool" layer. We replace the pooling layers of three different architectures of increasing complexity by CombPool layers, and empirically prove over multiple datasets that linear combinations outperform traditional pooling functions in most cases. Further, combinations with either the Sugeno integral or one of its generalizations usually yield the best results, proving a strong candidate to apply in most architectures.

Available data in machine learning applications is becoming increasingly complex, due to higher dimensionality and difficult classes. There exists a wide variety of approaches to measuring complexity of labeled data, according to class overlap, separability or boundary shapes, as well as group morphology. Many techniques can transform the data in order to find better features, but few focus on specifically reducing data complexity. Most data transformation methods mainly treat the dimensionality aspect, leaving aside the available information within class labels which can be useful when classes are somehow complex. This paper proposes an autoencoder-based approach to complexity reduction, using class labels in order to inform the loss function about the adequacy of the generated variables. This leads to three different new feature learners, Scorer, Skaler and Slicer. They are based on Fisher's discriminant ratio, the Kullback-Leibler divergence and least-squares support vector machines, respectively. They can be applied as a preprocessing stage for a binary classification problem. A thorough experimentation across a collection of 27 datasets and a range of complexity and classification metrics shows that class-informed autoencoders perform better than 4 other popular unsupervised feature extraction techniques, especially when the final objective is using the data for a classification task.

Evolutionary computation has largely exhibited its potential to complement conventional learning algorithms in a variety of machine learning tasks, especially those related to unsupervised (clustering) and supervised learning. It has not been until lately when the computational efficiency of evolutionary solvers has been put in prospective for training reinforcement learning models. However, most studies framed so far within this context have considered environments and tasks conceived in isolation, without any exchange of knowledge among related tasks. In this manuscript we present A-MFEA-RL, an adaptive version of the well-known MFEA algorithm whose search and inheritance operators are tailored for multitask reinforcement learning environments. Specifically, our approach includes crossover and inheritance mechanisms for refining the exchange of genetic material, which rely on the multilayered structure of modern deep-learning-based reinforcement learning models. In order to assess the performance of the proposed approach, we design an extensive experimental setup comprising multiple reinforcement learning environments of varying levels of complexity, over which the performance of A-MFEA-RL is compared to that furnished by alternative nonevolutionary multitask reinforcement learning approaches. As concluded from the discussion of the obtained results, A-MFEA-RL not only achieves competitive success rates over the simultaneously addressed tasks, but also fosters the exchange of knowledge among tasks that could be intuitively expected to keep a degree of synergistic relationship.

Social network group decision-making (SNGDM) develops rapidly because of the popularity of online connections among experts. The current SNGDM studies mainly focus on the influence of trust on the evolution of consensus opinions through a complex process. Some situations are ignored in the existing research when trust may negatively affect the decision-making process and trust relationships may be developed or interrupted during consensus negotiation processes. To overcome such limitations, we propose a consensus model based on the trust-consensus multiplex network by combing trust social network analysis and consensus evolution networks (CENs). We design an interaction mechanism between trust and consensus based on the dynamic experts’ influence which is computed by the multiplex PageRank centrality measure, especially focusing on the negative impact of trust on consensus. Besides, we compute consensus levels based on the density and intensity of CENs to determine when should the negotiation end. The proposed model can not only consider the mutual influence between trust and consensus in SNGDM, but also detect and analyze the negative influence of trust on consensus. An example examines the effectiveness of the proposed model and a comparative analysis shows its flexibility for studying the complex consensus situation of SNGDM.

In computing with words, it has been stressed that words mean different things for different people, which entails that decision makers (DMs) have personalized individual semantics (PISs) attached to linguistic expressions in linguistic group decision making (GDM). In particular, the PISs of DMs are not fixed, and they will be changing during the consensus building process, which indicates the necessary of continual PIS learning. Therefore, in this article, we propose a continual PIS-learning-based consensus approach in linguistic GDM. Specifically, a continual PIS learning model with the consistency-driven methodology is proposed to update the PISs taking into account all the linguistic preference data given by DMs during the consensus process. Then, the consensus measurement and feedback recommendation based on PIS are developed to detect the consensus process. Finally, numerical examples and simulation analysis are presented to illustrate and justify the use of the continual PIS-learning-based consensus approach.

In large-scale group decision making (LSGDM), appropriate clustering analysis is important to consensus reaching since it can reduce the interactive complexity among individuals. According to the traditional clustering method, a conflict may arise between the consensus reaching levels and total adjustment costs within clusters when individuals have different unit adjustment cost, which reflects their willingness to make concessions. Since this conflict may aggravate the consensus complexity, we propose a new$K$-means clustering method that considers both preferences and the preference adjustment cost. The preference adjustment cost is attached to preferences with a parameter that can be determined by balancing this conflict. Because of such conflict, the proposed clustering algorithm can improve the similarity of intracluster individuals on the preference adjustment cost by offsetting some acceptable consensus reaching levels within clusters. According to the proposed clustering algorithm, individuals who have both similar preferences and adjustment willingness are classified into the same clusters. In this way, the moderator can provide similar compensation strategies for intracluster individuals, which will decrease the adjustment complexity. A practical case study of team construction examines the application of the proposed algorithm, and the related comparative analysis shows that it is convenient for managers to persuade individuals to reach a consensus under the improved clustering results.

Ordinal regression addresses the problem of predicting non-numerical ordered classes. It walks a fine line between standard regression and classification, and the problem is often addressed from one of these perspectives. This can lead to suboptimal results as the ordinal information in the data may not be properly exploited. In this work we propose a distance metric learning algorithm to handle ordinal regression. Our model aims at optimizing the number of ordered sequences in local neighborhoods of the data, so that the learned distance can then be used by a distance-based predictor and improve its performance in ordinal regression problems. We evaluate our algorithm on several ordinal regression datasets and show that it outperforms the current distance metric learning for ordinal regression proposals, as well as being competitive with respect to the state-of-the-art of ordinal regression. The current paper is an extended abstract for the work [1].

Federated learning (FL) is an emerging and privacy-preserving machine learning technique that is shown to be increasingly important in the digital age. The two challenging issues for FL are: (1) communication overhead between clients and the server, and (2) volatile distribution of training data such as class imbalance. The paper aims to tackle these two challenges with the proposal of a federated fuzzy learning algorithm (FFLA) that can be used for data-based construction of fuzzy classification models in a distributed setting. The proposed learning algorithm is fast and highly cheap in communication by requiring only two rounds of interplay between the server and clients. Moreover, FFLA is empowered with an an imbalance adaptation mechanism so that it remains robust against heterogeneous distributions of data and class imbalance. The efficacy of the proposed learning method has been verified by the simulation tests made on a set of balanced and imbalanced benchmark data sets.

The Internet of Things (IoT) enables the interconnection of new cyber–physical devices that generate significant traffic of distributed, heterogeneous, and dynamic data at the network edge. Since several IoT applications demand for short response times (e.g., industrial applications, emergency management, real-time systems, and healthcare systems) and, at the same time, rely on resource-constrained devices, the adoption of traditional data mining techniques is neither effective nor efficient. Therefore, conventional data mining techniques need to be adjusted for optimizing response times, energy consumption, and data traffic while still providing adequate accuracy as required by the IoT applications.

Urban resettlement projects involve a large number of stakeholders and impose tremendous cost. Developing resettlement plans and reaching an agreement amongst stakeholders about resettlement plans at a reasonable cost are some of the key issues in urban resettlement. From this perspective, urban resettlement is a typical large-scale group decision-making (GDM) problem, which is challenging because of the scale of participants and the requirement of high consensus levels. Observing that residents who are affected by a resettlement project often have tight social connections, this study proposes a framework to improve the consensus reaching and uses the minimum consensus cost to reduce the total cost for urban resettlement projects with more than 1000 participants. Firstly, we construct a network topology that consists of two layers to deal with incomplete social relationships amongst large-scale participants. An inner layer consists of participants whose preference similarities and trust relations are known. Meanwhile, an outside layer includes participants whose trust relations cannot be determined. Secondly, we develop a classification method to classify participants into small subgroups based on their preference similarities. We can then connect the participants whose trust relations are unknown (the outside layer) with the ones in the inner layer using the classification results. To facilitate effective consensus reaching in large-scale social network GDM, we develop a three-step approach to reconcile conflicting preferences and accelerate the consensus process at the minimum cost. A real-life urban resettlement example is used to validate the proposed approach. Results show that the proposed approach can reduce the total consensus cost compared with the other two practices used in the actual urban resettlement operations.

Data in the real world is far from being perfect. The appearance of noise is a common issue that arises from the limitations of data acquisition mechanisms and human knowledge. In classification, label noise will hinder the performance of almost all classifiers, inducing a bias in the built model. While label noise has recently attracted researchers’ attention in standard classification, it has only recently begun to be studied in multiple instance classification. In this work, we propose the usage of filtering algorithms for multiple instance classification that are able to reduce the impact of negative instances within the bags. In order to do so, we decompose the bags to form a standard classification problem that can be efficiently treated by a specialized noise filter. Such a decomposition is tackled in different ways, with the aim of exploiting the knowledge offered by the examples from opposite bags. The bags are then rebuilt, without the identified noise instances. In our experiments, we show that by applying our approach we can diminish the impact of noise and even obtain better results at 0% noise level for several classifiers. Our approach sets out a promising approach to dealing with noise in the bags of multiple instance datasets and further improve the classification rate of the built models.

Data Science and Machine Learning have become fundamental assets for companies and research institutions alike. As one of its fields, supervised classification allows for class prediction of new samples, learning from given training data. However, some properties can cause datasets to be problematic to classify. In order to evaluate a dataset a priori, data complexity metrics have been used extensively. They provide information regarding different intrinsic characteristics of the data, which serve to evaluate classifier compatibility and a course of action that improves performance. However, most complexity metrics focus on just one characteristic of the data, which can be insufficient to properly evaluate the dataset towards the classifiers' performance. In fact, class overlap, a very detrimental feature for the classification process (especially when imbalance among class labels is also present) is hard to assess. This research work focuses on revisiting complexity metrics based on data morphology. In accordance to their nature, the premise is that they provide both good estimates for class overlap, and great correlations with the classification performance. For that purpose, a novel family of metrics have been developed. Being based on ball coverage by classes, they are named after Overlap Number of Balls. Finally, some prospects for the adaptation of the former family of metrics to singular (more complex) problems are discussed.

With the rapid development of society and continual progress of science and technology, large-scale group decision-making (LSGDM) problems are very commonly encountered in real-life situations. Considering that the information required for decision-making and people's cognition processes is becoming more and more complex, double hierarchy linguistic preference relation (DHLPR) can be used to express complex linguistic information reasonably and intuitively. Sometimes experts in LSGDM unwillingly modify their preferences or even modify them on purpose in a contrary way to the other experts. Thus, differing opinions or minority preferences are often referred to as obstacles to decision-making. This article develops a consensus model to manage minority opinions and noncooperative behaviors in LSGDM with DHLPRs. In addition, to establish the consensus model, some basic tools, such as the clustering method, weights-determining method, and adjustment coefficients-determining method, are developed. Finally, a practical LSGDM problem is set up to prove that the proposed consensus model is feasible and effective, and some comparative analyses are made to highlight the advantages of these methods and models, as well as to analyze current deficiencies.

When people express their opinions, they often cannot provide exact opinions but express uncertain opinions, such as numerical interval opinions. Moreover, due to the differences in the cultural background and character of agents, people who encounter numerical interval opinions often show different uncertainty tolerances. By taking different numerical interval opinions and different uncertainty tolerances into account, in this article, we propose a numerical interval opinion dynamics model to investigate the process of forming collective opinions in a group of interaction agents under an uncertain and social network context. We propose the theoretical analysis and algorithms to identify the stable agents whose opinions will be becoming stable and the oscillation agents whose opinions will always be fluctuating in the opinion evolution process. Furthermore, we study the conditions under which a consensus opinion can be built among the stable agents and estimate the opinion ranges of the oscillation agents. Finally, numerical examples and analysis are used to show the feasibility and effectiveness of the proposed theories and algorithms.

Opinion dynamics are investigated extensively to describe the process of opinion formation in groups of individuals. Most of the existing opinion dynamics models assume that the individuals express numerical opinions. However, this assumption does not consider the fact that people often express their opinions in a linguistic way. Particularly, for linguistic opinions, an important point to be highlighted in computing with words (CW) is that words mean different things to different people. In this article, following the idea of personalized individual semantics (PIS) model, we propose the PIS-based linguistic opinions dynamics model (PIS-LOD model) in the framework of bounded confidence effects. Then, some desired properties in the PIS-LOD model are discussed in detail. Furthermore, we design the detailed simulation experiments to show that the individuals' familiarity (which refers to the knowledge on others' semantics) and the PISs' differences have a great influence on the stabilized time, distributions of the extreme and moderate opinions, linguistic opinions distribution, semantics distribution, number of clusters, consensus reaching and the extremely small clusters. The results in this article are very helpful for us to understand the process of group opinion formation in a linguistic context.

Large-scale group decision-making (LSGDM) solution is usually based on the clustering analysis process (CAP) and consensus reaching process (CRP). However, CAP and CRP can be contradictory since CAP is performed based on the differences between potentially small groups and CRP is conducted to improve the overall similarity of a large group. To balance CAP and CRP, a dynamic clustering analysis process (DCAP) based on consensus evolution networks is proposed. A clustering algorithm proposed based on community detection method can be used to handle the diverse network structures with dynamic consensus thresholds. The clustering validity based on the intracluster consensus levels in subgroups and the intercluster consensus level among subgroups is evaluated. Then, the DCAP after each feedback adjustment round in CRP is reanalyzed. In such a way, effective clustering can also be found after a satisfying consensus is reached. Finally, a case study shows the availability of this approach and comparative analyses are provided to highlight the advantages from both theoretical and numerical perspectives.

To date, a large number of consensus reaching processes (CRPs) have been reported in group decision making (GDM). Trust relationships should be an essential element in interactions among a group of individuals, leading to the evolution of individuals’ preferences. Therefore, in this article, we present a trust relationships CRP with a feedback mechanism which consists of two approaches of facilitating consensus reaching: 1) the leader-based preference adjustment and 2) the trust relationships improvement. In the trust relationships CRP, we build a bridge between opinion dynamics and GDM to highlight the role of the leaders and trust relationships improvements in the GDM problems. Furthermore, we present a new strategic manipulation issue, called trust relationship manipulation, and discuss some clique-based strategies to manipulate trust relationships to obtain the desired ranking of the alternatives in the GDM problems. Finally, the detailed simulation experiments are proposed to justify our proposal.

This article provides a brief tour through the main fuzzy and linguistic decision-making trends, studies, methodologies, and models developed in the last 50 years. Fuzzy and linguistic decision-making approaches allow to address complex real-world decision problems where humans exhibit vagueness, imprecision, and/or use natural language to assess decision alternatives, criteria, etc. The aim of this article is threefold. First, the main fuzzy set theory and computing with words-based representation paradigms of decision information, with their different levels of expressive richness and complexity, are reviewed. Second, three core decision-making frameworks are examined: 1) multicriteria decision making; 2) group consensus-driven decision making; and 3) multiperson multicriteria decision making. Third, the article discusses new complex decision-making frameworks that have emerged in recent years, where decisions are guided by the “wisdom of the crowd”: their associated challenges are highlighted and considerations on much needed key guidelines for future research in the field are provided.

In recent years, Multifactorial optimization (MFO) has gained a notable momentum in the research community. MFO is known for its inherent capability to efficiently address multiple optimization tasks at the same time, while transferring information among such tasks to improve their convergence speed. On the other hand, the quantum leap made by Deep Q Learning (DQL) in the Machine Learning field has allowed facing Reinforcement Learning (RL) problems of unprecedented complexity. Unfortunately, complex DQL models usually find it difficult to converge to optimal policies due to the lack of exploration or sparse rewards. In order to overcome these drawbacks, pre-trained models are widely harnessed via Transfer Learning, extrapolating knowledge acquired in a source task to the target task. Besides, meta-heuristic optimization has been shown to reduce the lack of exploration of DQL models. This work proposes a MFO framework capable of simultaneously evolving several DQL models towards solving interrelated RL tasks. Specifically, our proposed framework blends together the benefits of meta-heuristic optimization, Transfer Learning and DQL to automate the process of knowledge transfer and policy learning of distributed RL agents. A thorough experimentation is presented and discussed so as to assess the performance of the framework, its comparison to the traditional methodology for Transfer Learning in terms of convergence, speed and policy quality, and the intertask relationships found and exploited over the search process.

Multitasking optimization is an incipient research area which is lately gaining a notable research momentum. Unlike traditional optimization paradigm that focuses on solving a single task at a time, multitasking addresses how multiple optimization problems can be tackled simultaneously by performing a single search process. The main objective to achieve this goal efficiently is to exploit synergies between the problems (tasks) to be optimized, helping each other via knowledge transfer (thereby being referred to as Transfer Optimization). Furthermore, the equally recent concept of Evolutionary Multitasking (EM) refers to multitasking environments adopting concepts from Evolutionary Computation as their inspiration for the simultaneous solving of the problems under consideration. As such, EM approaches such as the Multifactorial Evolutionary Algorithm (MFEA) has shown a remarkable success when dealing with multiple discrete, continuous, single-, and/or multi-objective optimization problems. In this work we propose a novel algorithmic scheme for Multifactorial Optimization scenarios - the Multifactorial Cellular Genetic Algorithm (MFCGA) - that hinges on concepts from Cellular Automata to implement mechanisms for exchanging knowledge among problems. We conduct an extensive performance analysis of the proposed MFCGA and compare it to the canonical MFEA under the same algorithmic conditions and over 15 different multitasking setups (encompassing different reference instances of the discrete Traveling Salesman Problem). A further contribution of this analysis beyond performance benchmarking is a quantitative examination of the genetic transferability among the problem instances, eliciting an empirical demonstration of the synergies emerged between the different optimization tasks along the MFCGA search process.

Agent-based modeling (ABM) is a simulation paradigm to model complex systems by defining heterogeneous individual-level behaviors in a bottom-up approach. ABM is typically employed to simulate markets to study consumer decisions and to see how consumers make their purchase decisions. In this work, we present a marketing ABM where consumer perceptions are modeled using 2-tuple fuzzy linguistic variables. These variables represent the opinions the consumers have on the different features of every product, which drive their decisions (e.g., price or quality). In contrast to numerical or crisp values, fuzzy linguistic variables are a realistic representation of these qualitative aspects. In our ABM, agents use a decision-making heuristic to select a product, which is based on those perceptions and a probabilistic utility maximization rule. This process requires a fuzzy aggregation of the perceptions of every product, based on an ordered weighted average (OWA). In addition, consumers can be aware or unaware of each product in the market. In our ABM, we model this information by introducing a brand awareness filter when applying the decision-making heuristic. Thus, consumer agents can only select those products they are aware of. Our experimental results show that our realistic representation of the consumer preferences is more accurate than other existing approaches.

When developing a Machine Learning model, the consideration of explainability as an additional design driver can improve its deployment into any application context. Given an audience, an explainable Artificial Intelligence system is one that produces details or reasons to make it's functioning clear or easy to understand. Among different paradigms that inherently support these capabilities, Fuzzy Rule Based Systems are a very accountable solution. The main issue when dealing with fuzzy systems is to select an appropriate granularity to represent (fuzzify) the input data. A low value may cause the generation of too generalist rules, causing a hinder on predictive performance, whereas a high value may lead to both overfitting and/or very complex solutions. To overcome this situation, we propose a novel hierarchical fuzzy classification system based on fuzzy exception rules. To do so, low granularity rules are first generated and their confidence is examined. For those cases in which the fuzzy confidence is below a quality threshold, new higher granularity rules are created to cover the "instances in conflict" for the general rule, which is still kept in the rule base. Experimental results show the achievement of a compact and interpretable final rule base while maintaining or improving the predictive performance in comparison with the baseline fuzzy rule based classification and hierarchical systems.

Detecting the dependency between integration test cases plays a vital role in the area of software test optimization. Classifying test cases into two main classes – dependent and independent – can be employed for several test optimization purposes such as parallel test execution, test automation, test case selection and prioritization, and test suite reduction. This task can be seen as an imbalanced classification problem due to the test cases’ distribution. Often the number of dependent and independent test cases is uneven, which is related to the testing level, testing environment and complexity of the system under test. In this study, we propose a novel methodology that consists of two main steps. Firstly, by using natural language processing we analyze the test cases’ specifications and turn them into a numeric vector. Secondly, by using the obtained data vectors, we classify each test case into a dependent or an independent class. We carry out a supervised learning approach using different methods for handling imbalanced datasets. The feasibility and possible generalization of the proposed methodology is evaluated in two industrial projects at Bombardier Transportation, Sweden, which indicates promising results.

Understanding consumer behaviors and how consumers react to marketing campaigns and viral word-of-mouth processes is crucial for marketers. Classical approaches try to infer this information from a global top-down perspective. However, a more suitable and natural approach is to model consumer behaviors in a heterogeneous and decentralized bottom-up approach. In this case, each virtual consumer has her own mental state and decision-making strategies to simulate her purchase decisions. The system of virtual consumers generates the global sales and a marketer can understand the rules that govern the market. A well-known paradigm to model these systems is agent-based modeling (ABM). In this manuscript we present an ABM where the brand preferences of the consumer agents are modeled using 2-tuple fuzzy linguistic variables. These variables represent the perceptions these consumers have on the different aspects or drivers every product available in the market has (e.g., price or quality). The product selection process of the agents is based on those perceptions and a utility maximization rule. This rule requires a fuzzy aggregation of the fuzzy linguistic perceptions about the products. Our proposal employs an ordered weighted average (OWA) to aggregate them. Our experiments show this approach does not suffer any loss of information when applied on data from real markets. Hence it is a suitable representation of the products preferences, normally represented by qualitative values in marketing surveys. To the best of our knowledge, this is the first work integrating a marketing ABM with fuzzy linguistic modeling.

The social network group decision-making is popular due to the advantages of social relationships in the consensus reaching process, especially the trust relationships. To explore the effects of trust on consensus, some minimum cost consensus models are proposed based on implicit trust between individuals and the moderator. The implicit trust is computed based on the similarity of opinions and it is implied into the traditional minimum cost consensus model to obtain a new quadratic programming problem and the related dual problem. The weights of individuals can be determined based on implicit trust and can be used to modify the possible deviations among individuals’ adjustment costs. A numerical example and the comparative analysis are given to analyze the effectiveness of the proposed models, which suggests that individuals are willing to give up some benefit to reach consensus due to their implicit trust to the moderator and make minor revisions to their adjustment costs due to their implicit trust to each other.

pyDML is an open-source python library that provides a wide range of distance metric learning algorithms. Distance metric learning can be useful to improve similarity learning algorithms, such as the nearest neighbors classifier, and also has other applications, like dimensionality reduction. The pyDML package currently provides more than 20 algorithms, which can be categorized, according to their purpose, in: dimensionality reduction algorithms, algorithms to improve nearest neighbors or nearest centroids classifiers, information theory based algorithms or kernel based algorithms, among others. In addition, the library also provides some utilities for the visualization of classifier regions, parameter tuning and a stats website with the performance of the implemented algorithms. The package relies on the scipy ecosystem, it is fully compatible with scikit-learn, and is distributed under GPLv3 license. Source code and documentation can be found at https://github.com/jlsuarezdiaz/pyDML.

In this paper, we present an interval MULTIMOORA method with complete interval computation in which the interval distances of interval numbers and preference matrix are used. In addition, we propose a group interval best-worst method (BWM) with interval preference degree. The group interval BWM has a hierarchical structure of group decision making with two levels of experts. Beside employing the dominance theory to integrate subordinate rankings, we introduce the interval Borda rule as an aggregation function which does not have the defects of the dominance theory. We calculate the objective interval weights of criteria based on the interval entropy method, which are integrated by the subjective weights computed by the group interval BWM. The approach presented in this paper is verified by a real-world engineering selection problem of a hybrid vehicle engine based on real data and opinions of engineering design experts of the automotive industry of Iran. The preference-based and dominance-based ranking lists are presented for the problem. We solve the same case by employing the interval TOPSIS and VIKOR methods. Eventually, all resultant rankings are compared based on Spearman rank correlation coefficients.

One of the best-known and most effective methods in supervised classification is the k-nearest neighbors algorithm (kNN). Several approaches have been proposed to improve its accuracy, where fuzzy approaches prove to be among the most successful, highlighting the classical fuzzy k-nearest neighbors (FkNN). However, these traditional algorithms fail to tackle the large amounts of data that are available today. There are multiple alternatives to enable kNN classification in big datasets, spotlighting the approximate version of kNN called hybrid spill tree. Nevertheless, the existing proposals of FkNN for big data problems are not fully scalable, because a high computational load is required to obtain the same behavior as the original FkNN algorithm. This article proposes global approximate hybrid spill tree FkNN and local hybrid spill tree FkNN, two approximate approaches that speed up runtime without losing quality in the classification process. The experimentation compares various FkNN approaches for big data with datasets of up to 11 million instances. The results show an improvement in runtime and accuracy over literature algorithms.

The eighteen papers in this special section focus on emerging techniques and applications supported by fuzzy logic and artificial intelligence (AI). AI has become the focus of the day and attracted much attention from researchers, industries, and governments. This special issue serves as a forum to bring together all emerging techniques for fuzzy logic and fuzzy set-based AI and foster new advancements along this important direction. Actually, there have been a number of research pursuits that position themselves at the junction of AI and fuzzy logic. For example, natural language processing, viewed as the jewel in the crown of AI, has been one of the focal points in the domain of fuzzy logic and fuzzy sets. Fuzzy sets can offer an effective paradigm supporting accurate understanding of natural language and build efficient linkages to human intelligence through concepts and computing with membership functions, in particular type-2 fuzzy sets for explainable AI.

Currently, Coronavirus disease (COVID-19), one of the most infectious diseases in the 21st century, is diagnosed using RT-PCR testing, CT scans and/or Chest X-Ray (CXR) images. CT (Computed Tomography) scanners and RT-PCR testing are not available in most medical centers and hence in many cases CXR images become the most time/cost effective tool for assisting clinicians in making decisions. Deep learning neural networks have a great potential for building COVID-19 triage systems and detecting COVID-19 patients, especially patients with low severity. Unfortunately, current databases do not allow building such systems as they are highly heterogeneous and biased towards severe cases. This article is three-fold: (i) we demystify the high sensitivities achieved by most recent COVID-19 classification models, (ii) under a close collaboration with Hospital Universitario Clínico San Cecilio, Granada, Spain, we built COVIDGR-1.0, a homogeneous and balanced database that includes all levels of severity, from normal with Positive RT-PCR, Mild, Moderate to Severe. COVIDGR-1.0 contains 426 positive and 426 negative PA (PosteroAnterior) CXR views and (iii) we propose COVID Smart Data based Network (COVID-SDNet) methodology for improving the generalization capacity of COVID-classification models. Our approach reaches good and stable results with an accuracy of [Formula: see text], [Formula: see text], [Formula: see text] in severe, moderate and mild COVID-19 severity levels. Our approach could help in the early detection of COVID-19. COVIDGR-1.0 along with the severity level labels are available to the scientific community through this link https://dasci.es/es/transferencia/open-data/covidgr/.

In this paper, we have proposed a new optimization algorithm, Memetic improved L-SHADE with a local search pool, MiLSHADE-LSP, a memetic algorithm that combines an improved L-SHADE with a local search pool. Improved L-SHADE modifies several important parameters during the run to encourage exploration in initial stages and to focus later the search around the most promising solutions. The local search pool is responsible to continuously improve the best solutions. MiLSHADE-LSP uses a pool of two different local search, LS, methods, the Broyden-Fletcher-Goldfarb-Shanno method with limited memory, L-BFGS-B, and the Solis-Wets algorithms, with an adaptive mechanism to choose which one of them is applied in each iteration selecting which had obtained a greater improvement last time it was applied. In order to avoid waste LS applications, the proposed algorithm stores a list of individuals that were not previously improved by each LS method. It also includes a restart mechanism to explore new areas when the search is stuck, restarting the population but maintaining the best found solution, and resetting the LS Pool parameters. In the experimental section we have tested and analyzed MiLSHADE-LSP using the proposed benchmark for the competition 100-digit challenge on Single Objective Numerical Optimization, obtaining that the LS Pool improves the algorithm, both achieving more optima and with a better performance. Results obtained show that MiLSHADE-LSP is a very competitive algorithm.

The state of the art in Sentiment Analysis is defined by deep learning methods, and currently the research efforts are focused on improving the encoding of underlying contextual information in a sequence of text.However, those neural networks with a higher representation capacity are increasingly more complex, which means that they have more hyper-parameters that have to be defined by hand.We argue that the setting of hyper-parameters may be defined as an optimisation task, we thus claim that evolutionary algorithms may be used to the optimisation of the hyper-parameters of a deep learning method.We propose the use of the evolutionary algorithm SHADE for the optimisation of the configuration of a deep learning model for the task of sentiment analysis in Twitter.We evaluate our proposal in a corpus of Spanish tweets, and the results show that the hyper-parameters found by the evolutionary algorithm enhance the performance of the deep learning method.

With the advent of Big Data, terabytes of data are generated and stored every second. This raw data is far from \nbeing perfect, it contains many imperfections (noise, missing values, etc.) and is not suitable for analysis, \nas it will led to wrong conclusions. Data preprocessing is the set of techniques devoted to polish, clean, \nfix, and improve that raw data. With this preprocessed data, we would be able to find more patterns in it, \nand to better explain the underlaying distribution of the data. This is what is called Smart Data, raw data \nthat has been preprocessed and is ready for being analyzed, data that contains valuable information that will \nled to knowledge. In this work, we present two Big Data libraries for achieving Smart Data from Big Data, \nBigDaPSpark and BigDaPFlink. They are built on top of two Big Data frameworks, Apache Spark and Apache \nFlink. Both libraries contain a series of algorithms for Big Data preprocessing, ranging from noise cleaning, \nto discretization, or data reduction, among many others. Additionally, we ilustrate the usage of the libraries \nwith two cases of use.

Albeit most techniques and algorithms assume that the data is accurate, measurements in our analogic world are far from being perfect. Since our capabilities of storing and processing data are growing everyday, these imperfections will accumulate, generating poorer decisions and hindering any knowledge extraction process carried out over the raw data. One of the most disturbing imperfections is the presence of missing values. Many inductive algorithms assume that the data is complete, thus if they face missing data they will not work properly or the quality of the knowledge extracted will be poorer. At this point there is no sophisticated missing values treatment implemented in any major Big Data framework. In this contribution, we present two novel imputation methods based on clustering that achieve better results than simply removing the faulty examples or filling-in the missing values with the mean that can be easily ported to Spark’s MLlib.

The quality of the data is directly related to the quality of the models drawn from that data. For that reason, many research is devoted to improve the quality of the data and to amend errors that it may contain. One of the most common problems is the presence of noise in classification tasks, where noise refers to the incorrect labeling of training instances. This problem is very disruptive, as it changes the decision boundaries of the problem. Big Data problems pose a new challenge in terms of quality data due to the massive and unsupervised accumulation of data. This Big Data scenario also brings new problems to classic data preprocessing algorithms, as they are not prepared for working with such amounts of data, and these algorithms are key to move from Big to Smart Data. In this paper, an iterative ensemble filter for removing noisy instances in Big Data scenarios is proposed. Experiments carried out in six Big Data datasets have shown that our noise filter outperforms the current state-of-the-art noise filter in Big Data domains. It has also proved to be an effective solution for transforming raw Big Data into Smart Data.

The current evolution in multidisciplinary learning analytics research poses significant challenges for the exploitation of behavior analysis by fusing data streams toward advanced decision-making. The identification of students that are at risk of withdrawals in higher education is connected to numerous educational policies, to enhance their competencies and skills through timely interventions by academia. Predicting student performance is a vital decision-making problem including data from various environment modules that can be fused into a homogenous vector to ascertain decision-making. This research study exploits a temporal sequential classification problem to predict early withdrawal of students, by tapping the power of actionable smart data in the form of students' interactional activities with the online educational system, using the freely available Open University Learning Analytics data set by employing deep long short-term memory (LSTM) model. The deployed LSTM model outperforms baseline logistic regression and artificial neural networks by 10.31% and 6.48% respectively with 97.25% learning accuracy, 92.79% precision, and 85.92% recall.

Preference relations have been widely used in group decision-making (GDM) problems. Recently, a new kind of preference relations called fuzzy preference relations with self-confidence (FPRs-SC) has been introduced, which allow experts to express multiple self-confidence levels when providing their preferences. This paper focuses on the analysis of additive consistency for FPRs-SC and its application in GDM problems. To do that, some operational laws for FPRs-SC are proposed. Subsequently, an additive consistency index that considers both the fuzzy preference values and self-confidence is presented to measure the consistency level of an FPR-SC. Moreover, an iterative algorithm that adjusts both the fuzzy preference values and self-confidence levels is proposed to repair the inconsistency of FPRs-SC. When an acceptable additive consistency level for FPRs-SC is achieved, the collective FPR-SC can be computed. We aggregate the individual FPRs-SC using a self-confidence indices-based induced ordered weighted averaging operator. The inherent rule for aggregation is to give more importance to the most self-confident experts. In addition, a self-confidence score function for FPRs-SC is designed to obtain the best alternative in GDM with FPRs-SC. Finally, the feasibility and validity of the research are demonstrated with an illustrative example and some comparative analyses.

Corals are crucial animals as they support a large part of marine life. The automatic classification of corals species based on underwater images is important as it can help experts to track and detect threatened and vulnerable coral species. However, this classification is complicated due to the nature of coral underwater images and the fact that current underwater coral datasets are unrealistic as they contain only texture images, while the images taken by autonomous underwater vehicles show the complete coral structure. The objective of this paper is two-fold. The first is to build a dataset that is representative of the problem of classifying underwater coral images, the StructureRSMAS dataset. The second is to build a classifier capable of resolving the real problem of classifying corals, based either on texture or structure images. We have achieved this by using a two-level classifier composed of three ResNet models. The first level recognizes whether the input image is a texture or a structure image. Then, the second level identifies the coral species. To do this, we have used a known texture dataset, RSMAS , and StructureRSMAS.

The combinatorial explosion problem is a great challenge for belief rule base (BRB) when a complex system has overnumbered attributes and/or referenced values for the attributes. This is because the BRB is conventionally constructed under the conjunctive assumption, conjunctive BRB, which requires covering each possible combination of all referenced values for all attributes. To solve this challenge, this study proposes a generic modeling, inferencing, and optimization approach for BRB under the disjunctive assumption, disjunctive BRB, that can significantly reduce its size. First, a disjunctive BRB is defined based on the mathematical description of the BRB space. The minimum size requirement for a disjunctive BRB is also discussed in comparison to a conjunctive one. Building on this, the generic disjunctive BRB modeling and inferencing procedures are proposed. Furthermore, an improved optimization model with further relaxed restrictions is constructed, and an optimization algorithm is developed in which only the new rule is optimized and its referenced values range is determined by the optimal solution in the former round optimization. The new optimization algorithm is more efficient with fewer variables and a more concise solution space. The results of three case studies confirm that by integrating both experts' knowledge and historic data, the modeling and inferencing processes can be well understood. Moreover, optimization can further improve the modeling accuracy while it facilitates downsizing the BRB in comparison with previous studies and other approaches.

There is a wide variety of studies that propose different classifiers to solve a large amount of problems in distinct classification scenarios. The no free lunch theorem states that if we use a big enough set of varied problems, all classifiers would be equivalent in performance. From another point of view, the performance of the classifiers is dependant of the scope and properties of the datasets. In this sense, new proposals on the topic often focus on a given context, aiming at improving the related state-of-the-art approaches. Data complexity metrics have been traditionally used to determine the inner characteristics of datasets. This way, researchers are able to categorize the problems in different scenarios. Then, this taxonomy can be applied to determine inner characteristics of the datasets in order to determine intervals of good and bad behavior for a given classifier. In this paper, we will take advantage of the data complexity metrics in order to design a fuzzy metaclassifier. The final goal is to create decision rules based on the inner characteristics of the data to apply a different version of the fuzzy classifier for a given problem. To do so, we will make use of the FARC-HD classifier, an evolutionary fuzzy system that has led to different extensions in the specialized literature. Experimental results show the goodness of this novel approach as it is able to outperform all versions of FARC-HD on a wide set of problems, and obtain competitive results (in terms of performance and interpretability) versus two selected state-of-the-art rule-based classification system, C4.5 and FURIA.

In this paper, a sparse representation-based intuitionistic fuzzy clustering (SRIFC) approach is presented for solving the large-scale decision making (LSDM) problem. It consists of two algorithms: the sparse representation-based intuitionistic fuzzy clustering-exactly precision algorithm (which is presented for an exactly precision requirement), and the sparse representation-based intuitionistic fuzzy clustering-soft precision and scalable algorithm (which is proposed for soft precision and scalable requirements). In the proposed SRIFC approach, decision makers are clustered into several interest groups according to their interest preferences and relation sparsity of their intuitionistic fuzzy assessment information. The purpose of the presented SRIFC approach is to investigate the group intra-relations among DMs and to detect the group leaders for each interest group during the clustering process. According to the illustrative experiment results, the presented SRIFC approach is an adaptive and the unsupervised clustering method and presents more robust and efficient for LSDM problems.

In linguistic large-scale group decision making (LSGDM), it is often necessary to achieve a consensus. Particularly, when computing with words and linguistic decision, we must keep in mind that words mean different things to different people. Therefore, to represent the specific semantics of each individual, we need to consider the personalized individual semantics (PIS) model in linguistic LSGDM. In this paper, we propose a consensus model based on PIS for LSGDM. Specifically, a PIS process to obtain the individual semantics of linguistic terms with linguistic preference relations is introduced. A consensus process based on PIS, including the consensus measure and feedback recommendation phases, is proposed to improve the willingness of decision makers who follow the suggestions to revise their preferences in order to achieve a consensus in linguistic LSGDM problems. The consensus measure defines two opposing consensus groups with respective acceptable and unacceptable consensus. In the feedback recommendation phase, a PIS-based clustering method to get decision makers with similar individual semantics is proposed. Recommendation rules design a feedback for decision makers with unacceptable consensus, finding suitable moderators from the decision makers with acceptable consensus based on cluster proximity.

The individual consistency and the consensus degree are two basic measures to conduct group decision making with reciprocal preference relations. The existing frameworks to manage individual consistency and consensus degree have been investigated intensively and follow a common resolution scheme composed by the two phases: the consistency improving process, and the consensus reaching process. But in these frameworks, the individual consistency will often be destroyed in the consensus reaching process, leading to repeat the consistency improving process, which is time consuming. In order to avoid repeating the consistency improving process, a consensus reaching process with individual consistency control is proposed in this paper. This novel consensus approach is based on the design of an optimization-based consensus rule, which can be used to determine the adjustment range of each preference value guaranteeing the individual consistency across the process. Finally, theoretical and numerical analysis are both used to justify the validity of our proposal.

Recently, large scale group decision making (LSGDM) problems have become a hotspot. This paper focuses on hesitant fuzzy LSGDM problems where decision makers (DMs) use hesitant fuzzy preference relations (HFPRs) to express their assessment information. HFPRs can represent the fuzziness and hesitancy of DM assessment information well. To improve the efficiency of hesitant fuzzy LSGDM problems, we first propose a reliability index-based consensus reaching process (RI-CRP). By assessing the ordinal consistency of DM's assessment information and measuring the deviation from collective opinion, the DM's opinion reliability index is given. To avoid unreliable information, we propose an unreliable DM management method to be used in the RI-CRP, based on the computation of DM's opinion reliability index. Moreover, an alternative ranking-based clustering (ARC) method with HFPRs is proposed to improve the efficiency of the RI-CRP. The similarity index between two DMs' opinions is provided to ensure the ARC method can be effectively implemented. Compared with those clustering methods which need to preset several correlated parameters, the presented ARC method is more objective with a different approach based on the alternative ranking. A numerical example shows that the proposed ARC method and the RI-CRP are feasible and effective for hesitant fuzzy LSGDM problems.

Designing efficient learning models capable of dealing with tons of data has become a reality in the era of big data. However, the amount of available data is too much for traditional data mining techniques to be applicable. This issue is even more serious when evolutionary algorithms are a key part of the learning algorithm. In this scenario, one typical approach is to follow a divide-and-conquer strategy, where data is divided into different chunks that are individually and independently addressed. Afterwards, the partial knowledge obtained from each chunk of data is combined in order to give a solution to the problem. Nevertheless, these kinds of local approaches do not look at data as a whole, missing a global view of the problem, which may result in less accurate models that also depend on how data is split. In this work, we focus on evolutionary feature selection algorithms. A divide-and-conquer approach to handle evolutionary feature selection in big data was already developed. We aim at designing its global counterpart, which looks at the feature selection problem from a global perspective, making use of the data as a whole to select the most appropriate features. In order to do so, we consider Apache Spark as a big data technology where our algorithm is implemented. We design a genetic algorithm capable of dealing with big datasets by selecting the proper parameters for our base algorithm (the well-known CHC) and adapting the evaluation procedure to take all the distributed data into account. Several preliminary results are discussed to study the feasibility of global evolutionary feature selection methods for big datasets.

Global optimization is a very important topic in research due to its wide applications in many real-world problems in science and engineering. Among optimization problems, dimensionality is one of the most crucial issues that increases the difficulty of the optimization process. Thus, Large-Scale Global Optimization, optimization with a great number of variables, arises as a field that is getting an increasing interest. In this paper, we propose a new hybrid algorithm especially designed to tackle this type of optimization problems. The proposal combines, in a iterative way, a modern Differential Evolution algorithm with one local search method chosen from a set of different search methods. The selection of the local search method is dynamic and takes into account the improvement obtained by each of them in the previous intensification phase, to identify the most adequate in each case for the problem. Experiments are carried out using the CEC'2013 Large-Scale Global Optimization benchmark, and the proposal is compared with other state-of-the-art algorithms, showing that the synergy among the different components of our proposal leads to better and more robust results than more complex algorithms. In particular, it improves the results of the current winner of previous Large-Scale Global Optimization competitions, Multiple Offspring Sampling, MOS, obtaining very good results, especially in the most difficult problems.

The Fuzzy k Nearest Neighbor (Fuzzy kNN) classifier is well known for its effectiveness in supervised learning problems. kNN classifies by comparing new incoming examples with a similarity function using the samples of the training set. The fuzzy version of the kNN accounts for the underlying uncertainty in the class labels, and it is composed of two different stages. The first one is responsible for calculating the fuzzy membership degree for each sample of the problem in order to obtain smoother boundaries between classes. The second stage classifies similarly to the standard kNN algorithm but uses the previously calculated class membership degree. To deal with very large datasets, distributed versions of the Fuzzy kNN algorithm have been proposed. However, existing approaches remain not fully scalable as they aim to replicate the exact behavior of the Fuzzy kNN. In this work, we present an approximate and distributed Fuzzy kNN approach based on Hybrid Spill-Tree implemented under Apache Spark. The aim of this model is to alleviate the scalability problems and to deal with big datasets maintaining high accuracy. In our experiments, we compare in precision and runtime with the Fuzzy kNN for big data problems existing in the literature, running with datasets of up to 11 million instances. The results show an improvement in the runtime and accuracy with respect to the previous exact model.

In recent years, evolutionary algorithms have been found to be effective and efficient techniques to train support vector machines (SVMs) for binary classification problems while multiclass problems have been neglected. This paper proposes CMOE-SVM: Cooperative Multi-Objective Evolutionary SVMs for multiclass problems. CMOE-SVM enables SVMs to handle multiclass problems via co-evolutionary optimization, by breaking down the original M-class problem into M simpler ones, which are optimized simultaneously in a cooperative manner. Furthermore, CMOE-SVM can explicitly maximize the margin and reduce the training error (the two components of the SVM optimization), by means of multi-objective optimization. Through a comprehensive experimental evaluation using a suite of benchmark datasets, we validate the performance of CMOE-SVM. The experimental results, supported by statistical tests, give evidence of the effectiveness of the proposed approach for solving multiclass classification problems.

BACKGROUND: The development of new ortholog detection algorithms and the improvement of existing ones are of major importance in functional genomics. We have previously introduced a successful supervised pairwise ortholog classification approach implemented in a big data platform that considered several pairwise protein features and the low ortholog pair ratios found between two annotated proteomes (Galpert, D et al., BioMed Research International, 2015). The supervised models were built and tested using a Saccharomycete yeast benchmark dataset proposed by Salichos and Rokas (2011). Despite several pairwise protein features being combined in a supervised big data approach; they all, to some extent were alignment-based features and the proposed algorithms were evaluated on a unique test set. Here, we aim to evaluate the impact of alignment-free features on the performance of supervised models implemented in the Spark big data platform for pairwise ortholog detection in several related yeast proteomes. RESULTS: The Spark Random Forest and Decision Trees with oversampling and undersampling techniques, and built with only alignment-based similarity measures or combined with several alignment-free pairwise protein features showed the highest classification performance for ortholog detection in three yeast proteome pairs. Although such supervised approaches outperformed traditional methods, there were no significant differences between the exclusive use of alignment-based similarity measures and their combination with alignment-free features, even within the twilight zone of the studied proteomes. Just when alignment-based and alignment-free features were combined in Spark Decision Trees with imbalance management, a higher success rate (98.71%) within the twilight zone could be achieved for a yeast proteome pair that underwent a whole genome duplication. The feature selection study showed that alignment-based features were top-ranked for the best classifiers while the runners-up were alignment-free features related to amino acid composition. CONCLUSIONS: The incorporation of alignment-free features in supervised big data models did not significantly improve ortholog detection in yeast proteomes regarding the classification qualities achieved with just alignment-based similarity measures. However, the similarity of their classification performance to that of traditional ortholog detection methods encourages the evaluation of other alignment-free protein pair descriptors in future research.

Fingerprint classification is one of the most common approaches to accelerate the identification in large databases of fingerprints. Fingerprints are grouped into disjoint classes, so that an input fingerprint is compared only with those belonging to the predicted class, reducing the penetration rate of the search. The classification procedure usually starts by the extraction of features from the fingerprint image, frequently based on visual characteristics. In this work, we propose an approach to fingerprint classification using convolutional neural networks, which avoid the necessity of an explicit feature extraction process by incorporating the image processing within the training of the classifier. Furthermore, such an approach is able to predict a class even for low-quality fingerprints that are rejected by commonly used algorithms, such as FingerCode. The study gives special importance to the robustness of the classification for different impressions of the same fingerprint, aiming to minimize the penetration in the database. In our experiments, convolutional neural networks yielded better accuracy and penetration rate than state-of-the-art classifiers based on explicit feature extraction. The tested networks also improved on the runtime, as a result of the joint optimization of both feature extraction and classification.

Computing with Words (CW) is a methodology which takes an essential role in decision making problems, thus the first aim of this paper is to highlight the importance of CW in decision making. There are many proposals of CW models that can be classified in two main fields: approaches based on membership functions and approaches based on qualitative scales. This paper focuses on the qualitative scales since it provides a closer environment to human knowledge and more accurate results avoiding a loss of information. An analysis and discussion of the qualitative scales of linguistic values in CW for decision making is provided. Finally, we highlight the open challenges for the qualitative scales in linguistic decision making.

The Synthetic Minority Oversampling Technique (SMOTE) preprocessing algorithm is considered "de facto" standard in the framework of learning from imbalanced data. This is due to its simplicity in the design of the procedure, as well as its robustness when applied to different type of problems. Since its publication in 2002, SMOTE has proven successful in a variety of applications from several different domains. SMOTE has also inspired several approaches to counter the issue of class imbalance, and has also significantly contributed to new supervised learning paradigms, including multilabel classification, incremental learning, semi-supervised learning, multi-instance learning, among others. It is standard benchmark for learning from imbalanced data. It is also featured in a number of different software packages - from open source to commercial. In this paper, marking the fifteen year anniversary of SMOTE, we reflect on the SMOTE journey, discuss the current state of affairs with SMOTE, its applications, and also identify the next set of challenges to extend SMOTE for Big Data problems.

Cloud Computing is an essential paradigm of computing services based on the “elasticity” property, where available resources are adapted efficiently to different workloads overtime. In elastic platforms, the forecasting component can be considered by far the most important element and the differentiating factor when comparing such systems, with workload forecasting one of the problems to solve if we want to achieve a truly elastic system. When properly addressed the cloud workload forecasting problem becomes a really interesting case study. As there is no general methodology in the literature that addresses this problem analytically and from a time series forecasting perspective (even less so in the cloud field), we propose a combination of these tools based on a state-of-the-art forecasting methodology which we have enhanced with some elements, such as: a specific cost function, statistical tests, visual analysis, etc. The insights obtained from this analysis are used to detect the asymmetrical nature of the forecasting problem and to find the best forecasting model from the viewpoint of the current state of the art in time series forecasting. From an operational point of view the most interesting forecast is a short-time horizon, so we focus on this. To show the feasibility of this methodology, we apply it to several realistic workload datasets from different datacenters. The results indicate that the analyzed series are non-linear in nature and that no seasonal patterns can be found. Moreover, on the analyzed datasets, the penalty cost as usually included in the SLA can be reduced to a 30 percent on average.

Since the analytic network process (ANP) is much more flexible than the analytic hierarchy process in handling the multiple criteria decision making (MCDM) problems in which the criteria or subcriteria are interdependent, it has attracted many scholars' attention and has been applied into many different areas. Given the powerfulness of intuitionistic fuzzy set in representing positive, negative, and indeterminate information, this paper investigates the ANP framework for the MCDM problems in which all the pairwise comparison judgment information over the objects are represented by intuitionistic fuzzy numbers. We first justify the way to decompose the MCDM problem into a holarchy and network structure, based on which, the intuitionistic fuzzy preference relations (IFPRs) can be constructed through pairwise comparisons over the goals, criteria, clusters as well as the elements. Considering that not all the IFPRs are consistent, we then propose a new method to derive the priorities from the IFPRs no matter the IFPRs are consistent or not. After that, we address the way to construct the supermatrix for those interdependent elements. The complete algorithm of intuitionistic fuzzy ANP (IFANP) is given and illustrated by a flow chart. To show the applicability and efficiency of the IFANP, we implement the method to a case study concerning the brand management of the six golden flowers of Sichuan liquor. Some comparative analyses are given to clarify the advantages and invalidation of the IFANP.

The hesitant fuzzy linguistic term set (HFLTS) is an effective tool to express the experts' subjective evaluations in the processes of decision making. To solve the problem with both qualitative and quantitative criteria in the context of HFLTSs and the crisp weights of criteria being unknown, this paper proposes a new multicriteria decision making method. First, formulas are developed to convert the quantitative data into the hesitant fuzzy linguistic elements. Then, motivated by the ORESTE method, we develop a new global preference score function to aggregate the criterion weights and criterion values, both of which are expressed as hesitant fuzzy linguistic elements. To get the real relation between alternatives, three preference intensity formulas are proposed and a hesitant fuzzy linguistic indifference threshold is introduced. We establish a conflict test framework after detailed research on the threshold values. On these bases, a new hesitant fuzzy linguistic ORESTE method is developed and the calculation process of this method is described. A case study on supplier selection is then presented to illustrate the method. Finally, some comparative analyses with other methods are conducted to show the practicability and reliability of the proposed method.

The probabilistic linguistic term set (PLTS) is a powerful technique in representing linguistic evaluations of individuals or groups in the process of decision making. The aim of this paper is to propose a strongly robust method to solve multiexperts multicriteria decision making problems with linguistic evaluations. To enrich the computation and to improve the measures of PLTS, we first define an expectation function of it. In addition, we advance three kinds of probabilistic linguistic distance measures reflecting on the difference of linguistic terms and probabilities at the same time to make up for the defects of the existing distance measures, and then propose the similarity and correlation measures. Integrating the subjective opinions with the correlation coefficients between criteria, we put forward a combined weight determining method. The robustness of the ranking method, MULTIMOORA, is enhanced by the improved Borda rule. Based on these research findings, a probabilistic linguistic MULTIMOORA method is proposed. Finally, the developed method is applied to an empirical example concerning the selection of shared karaoke television brands. The effectiveness of the proposed method is verified by some comparative analyses.

With the advent of extremely high dimensional datasets, dimensionality reduction techniques are becoming mandatory. Of the many techniques available, feature selection (FS) is of growing interest for its ability to identify both relevant features and frequently repeated instances in huge datasets. We aim to demonstrate that standard FS methods can be parallelized in big data platforms like Apache Spark so as to boost both performance and accuracy. We propose a distributed implementation of a generic FS framework that includes a broad group of well-known information theory-based methods. Experimental results for a broad set of real-world datasets show that our distributed framework is capable of rapidly dealing with ultrahigh-dimensional datasets as well as those with a huge number of samples, outperforming the sequential version in all the cases studied.

For years, there have been organized single objective real-parameter optimization competitions on the IEEE Congress on Evolutionary Computation, in which the organizer define a common experimental, the researchers carry out the experiments with their proposals using it, and the obtained results are compared. It is a excellent way to know which algorithms (and ideas) can improve others, creating guidelines to improve the field. However, in several competitions the benchmark can change and the winners of previous benchmarks are not always introduced into the comparisons. Due to that, it could be not clear the improvement that new proposals offer against proposals of previous years. In this paper, we compare the winners in different years among them using the different proposed benchmarks, and we analyse the results obtained by all of them to observe whether there is an real improvement or not by the winner proposals of these competitions through the years.

The design of efficient big data learning models has become a common need in a great number of applications. The massive amounts of available data may hinder the use of traditional data mining techniques, especially when evolutionary algorithms are involved as a key step. Existing solutions typically follow a divide-and-conquer approach in which the data is split into several chunks that are addressed individually. Next, the partial knowledge acquired from every slice of data is aggregated in multiple ways to solve the entire problem. However, these approaches are missing a global view of the data as a whole, which may result in less accurate models. In this work we carry out a first attempt on the design of a global evolutionary undersampling model for imbalanced classification problems. These are characterised by having a highly skewed distribution of classes in which evolutionary models are being used to balance the dataset by selecting only the most relevant data. Using Apache Spark as big data technology, we have introduced a number of variations to the well-known CHC algorithm to work with very large chromosomes and reduce the costs associated to the fitness evaluation. We discuss some preliminary results, showing the great potential of this new kind of evolutionary big data model.

The significance and benefits of addressing classification tasks in Big Data applications is beyond any doubt. To do so, learning algorithms must be scalable to cope with such a high volume of data. The most suitable option to reach this objective is by using a MapReduce programming scheme, in which algorithms are automatically executed in a distributed and fault tolerant way. Among different available tools that support this framework, Spark has emerged as a “de facto” solution when using iterative approaches. In this work, our goal is to design and implement an Evolutionary Fuzzy Rule Selection algorithm within a Spark environment. To do so, we build different local rule bases within each Map Task that are later optimized by means of a genetic process. With this procedure, we seek to minimize the total number of rules that are gathered by each Reduce task to obtain a compact and accurate Fuzzy Rule Based Classification System. In particular, we set the experimental framework in the scenario of imbalanced classification. Therefore, the final objective will be analyzing the best synergy between the novel Evolutionary Fuzzy Rule Selection algorithm and the solutions applied to cope with skewed class distributions, namely cost-sensitive learning, random under-sampling and random-oversampling.

In decision making dealing with computing with words, the importance of the statement that words mean different things for different people has been highlighted. In this paper, we focus on personalizing numerical scales of linguistic terms in decision making with hesitant fuzzy linguistic preference relations (HFLPRs). First, an average consistency measure for HFLPRs is provided, and then an optimization-based model to personalize individual semantics via numerical scales is presented, aiming at maximizing the average consistency of HFLPRs. Numerical examples are used to illustrate the proposal.

The k-Nearest Neighbors (kNN) classifier is one of the most effective methods in supervised learning problems. It classifies unseen cases comparing their similarity with the training data. Nevertheless, it gives to each labeled sample the same importance to classify. There are several approaches to enhance its precision, with the Fuzzy k-Nearest Neighbors (Fuzzy-kNN) classifier being among the most successful ones. Fuzzy-kNN computes a fuzzy degree of membership of each instance to the classes of the problem. As a result, it generates smoother borders between classes. Apart from the existing kNN approach to handle big datasets, there is not a fuzzy variant to manage that volume of data. Nevertheless, calculating this class membership adds an extra computational cost becoming even less scalable to tackle large datasets because of memory needs and high runtime. In this work, we present an exact and distributed approach to run the Fuzzy-kNN classifier on big datasets based on Spark, which provides the same precision than the original algorithm. It presents two separately stages. The first stage transforms the training set adding the class membership degrees. The second stage classifies with the kNN algorithm the test set using the class membership computed previously. In our experiments, we study the scaling-up capabilities of the proposed approach with datasets up to 11 million instances, showing promising results.

We present Teranga Go!, a social network with a linguistic fuzzy model which deals with HFLTS information as a practical application of decision making problems. It is defined to help members to select to whom interact based on collective information regarding real interactions with any user. In this way, we provide a tool intended to build trust among members of a sharing economy community given that is a major drawback from online transactions. As a workbench to run the linguistic decision making model, a web site and a mobile application for iOS and Android offer access to a carpooling service named Teranga Go! that seek to foster the mobility of international migration flows from Europe to Africa, based on concepts of collaborative economy and participatory consumption. The novelty of the site is the possibility of using hesitant linguistic expressions to assess a set of qualitative criteria and the use of the community members as the pool of experts. Unlike many multi criteria decision making problems we do not rank alternatives, we just qualify them using the retrieved opinions, which target a given user, and are collected over any interaction with this person along the time. Based on Computing with Words methodology where inputs are words and output are also words, we obtain from the model a linguistic value that is used to represent a karma property present in the user profile.

This paper proposes a multilabel fuzzy decision tree classifier named FuzzDTML. The algorithm uses generalized fuzzy entropy, aggregated over all labels, to choose the best attribute for growing the tree. The proposed algorithm also can generate leaves predicting partial label sets, which can incorporate to some degree the dependence among labels, as well as produce more interpretable models. An empirical analysis shows that, although the algorithm does not yet incorporate pruning nor fuzzy interval adjustment phases, it is competitive with other tree based approaches for multilabel classification, with better performance in data sets having numerical features that can be fuzzified.

In the Sentiment Analysis area, several Sentiment Analysis Methods are able to extract different polarities (positivity, neutrality and negativity) from written text, although neutrality is not usually considered because it is perceived as an insightfulness polarity. As it is known, the vote of majority has a high impact in decision making. We propose the use of Induced Ordered Weighted Averaging operators based on fuzzy majority for aggregating polarities from several Sentiment Analysis Methods. The main contribution of this work is to set neutrality on opinions guided by fuzzy majority. We aim to improve classification results by removing those neutral reviews labelled by a consensus of a collection.

With the advent of large-scale problems, feature selection has become a fundamental preprocessing step to reduce input dimensionality. The minimum-redundancy-maximum-relevance (mRMR) selector is considered one of the most relevant methods for dimensionality reduction due to its high accuracy. However, it is a computationally expensive technique, sharply affected by the number of features. This paper presents fast-mRMR, an extension of mRMR, which tries to overcome this computational burden. Associated with fast-mRMR, we include a package with three implementations of this algorithm in several platforms, namely, CPU for sequential execution, GPU (graphics processing units) for parallel computing, and Apache Spark for distributed computing using big data technologies.

Imbalanced classification is related to those problems that have an uneven distribution among classes. In addition to the former, when instances are located into the overlapped areas, the correct modeling of the problem becomes harder. Current solutions for both issues are often focused on the binary case study, as multi-class datasets require an additional effort to be addressed. In this research, we overcome these problems by carrying out a combination between feature and instance selections. Feature selection will allow simplifying the overlapping areas easing the generation of rules to distinguish among the classes. Selection of instances from all classes will address the imbalance itself by finding the most appropriate class distribution for the learning task, as well as possibly removing noise and difficult borderline examples. For the sake of obtaining an optimal joint set of features and instances, we embedded the searching for both parameters in a Multi-Objective Evolutionary Algorithm, using the C4.5 decision tree as baseline classifier in this wrapper approach. The multi-objective scheme allows taking a double advantage: the search space becomes broader, and we may provide a set of different solutions in order to build an ensemble of classifiers. This proposal has been contrasted versus several state-of-the-art solutions on imbalanced classification showing excellent results in both binary and multi-class problems.

Support vector machines (SVMs) are among the most powerful learning algorithms for classification tasks. However, these algorithms require a high computational cost during the training phase, which can limit their application on large-scale datasets. Moreover, it is known that their effectiveness highly depends on the hyper-parameters used to train the model. With the intention of dealing with these, this paper introduces an evolutionary multiobjective model and instance selection (IS) approach for SVMs with Pareto-based ensemble, whose goals are, precisely, to optimize the size of the training set and the classification performance attained by the selection of the instances, which can be done using either a wrapper or a filter approach. Due to the nature of multiobjective evolutionary algorithms, several Pareto optimal solutions can be found. We study several ways of using such information to perform a classification task. To accomplish this, our proposal performs a processing over the Pareto solutions in order to combine them into a single ensemble. This is done in five different ways, which are based on: 1) a global Pareto ensemble; 2) error reduction; 3) a complementary error reduction; 4) maximized margin distance; and 5) boosting. Through a comprehensive experimental study we evaluate the suitability of the proposed approach and the Pareto processing, and we show its advantages over a single-objective formulation, traditional IS techniques, and learning algorithms.

The papers in this special section address the use of fuzzy systems in data science. Fuzzy Systems in Data Science data science employs theories and techniques drawn from many fields to achieve knowledge extraction from large volumes of data. To facilitate and automate the achievement of useful insights, predictions, and decisions from collected data sets, scientists and engineers turned to Machine Learning solutions. Currently, with the birth of Big Data, the set of opportunities for inquiry has grown exponentially, thanks to these large and complex data sets. Being able to exploit this massive data effectively provides useful knowledge for decision-making or the exploration and comprehension of the phenomenon that produced the data. Data science includes several problems and tasks depending on the nature of the data and the type of knowledge that is to be extracted. From predictive and descriptive analysis to the application of recommendation systems, scientists and practitioners have considered the use of Soft Computing techniques based on fuzzy sets to overcome their problems. Fuzzy approaches have been applied with much success in several areas related to Data Science. Fuzzy sets can be used to effectively describe and incorporate uncertain data values, data measurement, and data relations; fuzzy systems can directly aid reasoning and inferencing in a learning machine and can be combined with machine learning methodologies to model human behaviors and complex data systems.

Mining massive and high-speed data streams among the main contemporary challenges in machine learning. This calls for methods displaying a high computational efficacy, with ability to continuously update their structure and handle ever-arriving big number of instances. In this paper, we present a new incremental and distributed classifier based on the popular nearest neighbor algorithm, adapted to such a demanding scenario. This method, implemented in Apache Spark, includes a distributed metric-space ordering to perform faster searches. Additionally, we propose an efficient incremental instance selection method for massive data streams that continuously update and remove outdated examples from the case-base. This alleviates the high computational requirements of the original classifier, thus making it suitable for the considered problem. Experimental study conducted on a set of real-life massive data streams proves the usefulness of the proposed solution and shows that we are able to provide the first efficient nearest neighbor solution for high-speed big and streaming data.

This paper presents a new version of the Coral Reefs Optimization (CRO) algorithm to improve its performance in large scale global optimization problems. Specifically, we propose to extend the original CRO with different substrate layers, where several exploration operators are defined. This definition allows establishing a competitive co-evolution process within the CRO, which improves the search for optimal solution in large scale optimization problems. The new CRO with substrate layers (CRO-SL) is used in combination with a local search, and the final memetic algorithm obtained has been tested by using a test suite for large scale continuous optimization, showing a robust behavior.

The classification of datasets with a skewed class distribution is an important problem in data mining. Evolutionary undersampling of the majority class has proved to be a successful approach to tackle this issue. Such a challenging task may become even more difficult when the number of the majority class examples is very big. In this scenario, the use of the evolutionary model becomes unpractical due to the memory and time constrictions. Divide-and-conquer approaches based on the MapReduce paradigm have already been proposed to handle this type of problems by dividing data into multiple subsets. However, in extremely imbalanced cases, these models may suffer from a lack of density from the minority class in the subsets considered. Aiming at addressing this problem, in this contribution we provide a new big data scheme based on the new emerging technology Apache Spark to tackle highly imbalanced datasets. We take advantage of its in-memory operations to diminish the effect of the small sample size. The key point of this proposal lies in the independent management of majority and minority class examples, allowing us to keep a higher number of minority class examples in each subset. In our experiments, we analyze the proposed model with several data sets with up to 17 million instances. The results show the goodness of this evolutionary undersampling model for extremely imbalanced big data classification.

Harmonic filtering has been widely applied to reduce harmonic distortion in power distribution systems. This paper investigates a new method of exploiting Differential Evolution (DE) to support the optimal design of harmonic filters. DE is a class of stochastic and population-based optimization algorithms that are expected to have stronger global ability than trajectory-based optimization techniques in locating the best component sizes for filters. However, the performance of DE is largely affected by its two control parameters: scaling factor and crossover rate, which are problem dependent. How to decide appropriate setting for these two parameters presents a practical difficulty in real applications. Greedy Adaptive Differential Evolution (GADE) algorithm is suggested in the paper as a more convenient and effective means to automatically optimize filter designs. GADE is attractive in that it does not require proper setting of the scaling factor and crossover rate prior to the running of the program. Instead it enables dynamic adjustment of the DE parameters during the course of search for performance improvement. The results of tests on several problem examples have demonstrated that the use of GADE leads to the discovery of better filter circuits facilitating less harmonic distortion than the basic DE method.

The treatment and processing of Big Data problems imply an essential advantage for researchers and corporations. This is due to the huge quantity of knowledge that is hidden within the vast amount of information that is available nowadays. In order to be able to address with such volume of information in an efficient way, the scalability for Big Data applications is achieved by means of the MapReduce programming model. It is designed to divide the data into several chunks or groups that are processed in parallel, and whose result is “assembled” to provide a single solution.

The study of consistency is a very important problem in decision making using preference relations. This paper focuses on measuring the consistency of hesitant fuzzy linguistic preference relations (HFLPRs). In this paper we propose the optimization-based approach to estimate the range of consistency degree in a HFLPR. The underlying idea of the proposed approach consists in measuring the pessimistic consistency index (PCI) of HFLPRs, and also the optimistic consistency index (OCI) of HFLPRs. The PCI of HFLPRs is determined by its linguistic preference relation with the worst consistency degree, and the OCI of HFLPRs is determined by its linguistic preference relation with the best consistency degree. Furthermore, numerical examples are provided to show the use of the proposed consistency measure.

This contribution deals with multi-instance classification, where the labeled data samples are bags composed on instances instead of labeled instances as in standard classification. Every bag contains a number of traditional instances (described by a number of attributes) and the number of instances is not usually the same in all the bags. So, the whole bag is labeled but the instances that compose the bag are not individually labeled. We propose a fuzzy sets based extension of the well known algorithm called Citation-KNN, a reference method in multi-instance classification. Citation-KNN uses two types of examples in the classification rule: neighbors and citers of the bag to be classified. We analyze two versions of our proposal, one of them using both neighbors and citers, and the other one using only neighbors. Our approach uses the Hausdorff distance and it is based on the FuzzyKNN algorithm. Several data-sets from KEEL data-set repository are used in the experimental study and we compare our proposals with the original Citation-KNN algorithm.

Discretization is one of the most relevant techniques for data preprocessing. The main goal of discretization is to transform numerical attributes into discrete ones to help the experts to understand the data more easily, and it also provides the possibility to use some learning algorithms which require discrete data as input, such as Bayesian or rule learning. We focus our attention on handling multivariate classification problems, where high interactions among multiple attributes exist. In this paper, we propose the use of evolutionary algorithms to select a subset of cut points that defines the best possible discretization scheme of a data set using a wrapper fitness function. We also incorporate a reduction mechanism to successfully manage the multivariate approach on large data sets. Our method has been compared with the best state-of-the-art discretizers on 45 real datasets. The experiments show that our proposed algorithm overcomes the rest of the methods producing competitive discretization schemes in terms of accuracy, for C4.5, Naive Bayes, PART, and PrUning and BuiLding Integrated in Classification classifiers; and obtained far simpler solutions.

In this paper, we review the definition and basic properties of the different types of fuzzy sets that have appeared up to now in the literature. We also analyze the relationships between them and enumerate some of the applications in which they have been used.

This letter makes some observations about “Interval type-2 fuzzy sets are generalization of interval-valued fuzzy sets: Towards a wide view on their relationship,”IEEE Trans. Fuzzy Systemsthat further support the distinction between an interval type-2 fuzzy set (IT2 FS) and an interval-valued fuzzy set (IV FS), points out that all operations, methods, and systems that have been developed and published about IT2 FSs are, so far, only valid in the special case when IT2 FS = IVFS, and suggests some research opportunities.

Multi-instance learning is a setting in supervised learning where the data consist of bags of instances. Samples in the dataset are groups of individual instances. In classification problems, a decision value is assigned to the entire bag, and the classification of an unseen bag involves the prediction of the decision value based on the instances it contains. In this paper, we develop a framework for multi-instance classifiers based on fuzzy set theory. Fuzzy sets have been used in many machine learning applications, but so far not in the classification of multi-instance data. We explore its untapped potential here. We interpret the classes as fuzzy sets and determine membership degrees of unseen bags to these sets based on the available training data. In doing so, we develop a framework of classifiers that extract the required membership degrees either at the level of instances (instance-based) or at the level of bags (bag-based). We offer an extensive analysis of the different settings within the proposed framework. We experimentally compare our proposal to state-of-the-art multi-instance classifiers, and based on two evaluation measures, our methods are shown to perform very well.

No abstract available.

Continuous optimization is an important research field because many real-world problems from very different domains (biology, engineering, data mining, etc.) can be formulated as the optimization of a continuous function. Into continuous optimization, solving high-dimensional optimization problems, also called large scale optimization, is a difficult challenge by the huge expansion of the domain search with the dimensionality. In this paper we propose a new hybrid algorithm to tackle this type of optimization, combining a DE with a LS method in a iterative way. The sharing of the best solution between these components in combination with a memory making possible a more in-depth search in each one of them, allowing the algorithm to obtain good results. Experiments are carried out using a benchmark designed for large scale optimization, and the proposal is compared with other algorithms, showing that the algorithm is robust, obtaining good results specially in the most difficult functions.

Classification techniques in the big data scenario are in high demand in a wide variety of applications. The huge increment of available data may limit the applicability of most of the standard techniques. This problem becomes even more difficult when the class distribution is skewed, the topic known as imbalanced big data classification. Evolutionary undersampling techniques have shown to be a very promising solution to deal with the class imbalance problem. However, their practical application is limited to problems with no more than tens of thousands of instances. In this contribution we design a parallel model to enable evolutionary undersampling methods to deal with large-scale problems. To do this, we rely on a MapReduce scheme that distributes the functioning of these kinds of algorithms in a cluster of computing elements. Moreover, we develop a windowing approach for class imbalance data in order to speed up the undersampling process without losing accuracy. In our experiments we test the capabilities of the proposed scheme with several data sets with up to 4 million instances. The results show promising scalability abilities for evolutionary undersampling within the proposed framework.

The advantages of multi-classification schemes based on decomposition strategies, and especially the One-vs-One framework, have been stressed even for those algorithms that can address multiple classes. However, there is an inherent hitch for the One-vs-One learning scheme related to the decision process: the non-competent classifier problem. This issue refers to the case where a binary classifier outputs a score degree for a couple of classes that are not related with the input example, thus including “noise” in the score-matrix and degrading the final accuracy. For this reason, several approaches have been developed in order to address the influence of the non-competence. Among them, the distance-based combination strategy has excelled as a very robust solution. In this contribution, we aim at investigating the behaviour of this approach using Evolutionary Fuzzy Systems as baseline classifiers. We will show that the synergy between both methodologies allows a significant improvement of the results to be obtained in contrast to the standard classifier and the classical One-vs-One scheme.

Decision making has become a core research area in different fields such as evaluation, engineering, medicine, etc. Usually, decision making problems are defined in contexts with vague and imprecise information. The use of linguistic modeling has provided successful results in decision making problems. However, most of the linguistic approaches are limited, because they restrict the elicitation of linguistic information to single linguistic terms and sometimes due to the lack of information, time or knowledge, decision makers hesitate among several linguistic terms to elicit their assessments and the use of only one linguistic term cannot reflect their assessments in a proper way. Therefore, more elaborated expressions than single linguistic terms might support decision makers in such hesitant situations and improve the elicitation of hesitant linguistic information. The use of hesitant fuzzy linguistic term sets (HFLTS), allows modeling this hesitation and facilitates the generation of comparative linguistic expressions similar to the expressions used by human beings in real world decision making problems using context-free grammars. There are different decision making models that deal with HFLTS, however they do not provide linguistic results as the computing with words scheme proposed to facilitate their comprehension. Therefore, the aim of this contribution is to present a multicriteria decision making model that not only improves the elicitation of hesitant linguistic information, but also obtains linguistic results easy to understand by decision makers. To achieve this latter goal the proposed model will make use of the linguistic 2-tuple model.

This contribution proposes a genetic learning process for designing the knowledge base of Fuzzy Rule-Based classification Systems, that will be used as binary classifiers in a One-vs-One decomposition for multi-class problems. A Genetic Algorithm is designed to adapt the number of fuzzy labels per variable (granularity level) for each classifier in order to improve the accuracy rate of a multi-class classifier. The genetic learning process evolves granularity levels and needs a fuzzy rules generation method for generating the whole knowledge base of the Fuzzy System. Several data-sets from KEEL data-set repository are used in the experimental study and we compare our proposal with three related methods: the standard way to design Fuzzy Rule-Based Classification Systems using the fuzzy rules generation method chosen with and without One-vs-One decomposition, and our proposal of genetic granularity level learning without One-vs-One decomposition.

Noise filters are preprocessing techniques designed to improve data quality in classification tasks by detecting and eliminating examples that contain errors or noise. However, filtering can also remove correct examples and examples containing valuable information, which could be useful for learning. This fact usually implies a margin of improvement on the noise detection accuracy for almost any noise filter. This paper proposes a scheme to improve the performance of noise filters in multi-class classification problems, based on decomposing the dataset into multiple binary subproblems. Decomposition strategies have proven to be successful in improving classification performance in multi-class problems by generating simpler binary subproblems. Similarly, we adapt the principles of the One-vs-One decomposition strategy to noise filtering, making the noise identification process simpler. In order to integrate the filtering results achieved in the binary subproblems, our proposal uses a soft voting approach considering a reliability level based on the aggregation of the noise degree prediction calculated for each binary classifier. The experimental results show that the One-vs-One decomposition strategy usually increases the performance of the noise filters studied, which can detect more accurately the noisy examples.

Self-labeled techniques are semi-supervised classification methods that address the shortage of labeled examples via a self-learning process based on supervised models. They progressively classify unlabeled data and use them to modify the hypothesis learned from labeled samples. Most relevant proposals are currently inspired by boosting schemes to iteratively enlarge the labeled set. Despite their effectiveness, these methods are constrained by the number of labeled examples and their distribution, which in many cases is sparse and scattered. The aim of this paper is to design a framework, named synthetic examples generation for self-labeled semi-supervised classification, to improve the classification performance of any given self-labeled method by using synthetic labeled data. These are generated via an oversampling technique and a positioning adjustment model that use both labeled and unlabeled examples as reference. Next, these examples are incorporated in the main stages of the self-labeling process. The principal aspects of the proposed framework are: 1) introducing diversity to the multiple classifiers used by using more (new) labeled data; 2) fulfilling labeled data distribution with the aid of unlabeled data; and 3) being applicable to any kind of self-labeled method. In our empirical studies, we have applied this scheme to four recent self-labeled methods, testing their capabilities with a large number of data sets. We show that this framework significantly improves the classification capabilities of self-labeled techniques.

There are many real-world classification problems involving multiple classes, e.g., in bioinformatics, computer vision, or medicine. These problems are generally more difficult than their binary counterparts. In this scenario, decomposition strategies usually improve the performance of classifiers. Hence, in this paper, we aim to improve the behavior of fuzzy association rule-based classification model for high-dimensional problems (FARC-HD) fuzzy classifier in multiclass classification problems using decomposition strategies, and more specifically One-versus-One (OVO) and One-versus-All (OVA) strategies. However, when these strategies are applied on FARC-HD, a problem emerges due to the low-confidence values provided by the fuzzy reasoning method. This undesirable condition comes from the application of the product t-norm when computing the matching and association degrees, obtaining low values, which are also dependent on the number of antecedents of the fuzzy rules. As a result, robust aggregation strategies in OVO, such as the weighted voting obtain poor results with this fuzzy classifier. In order to solve these problems, we propose to adapt the inference system of FARC-HD replacing the product t-norm with overlap functions. To do so, we define n-dimensional overlap functions. The usage of these new functions allows one to obtain more adequate outputs from the base classifiers for the subsequent aggregation in OVO and OVA schemes. Furthermore, we propose a new aggregation strategy for OVO to deal with the problem of the weighted voting derived from the inappropriate confidences provided by FARC-HD for this aggregation method. The quality of our new approach is analyzed using 20 datasets and the conclusions are supported by a proper statistical analysis. In order to check the usefulness of our proposal, we carry out a comparison against some of the state-of-the-art fuzzy classifiers. Experimental results show the competitiveness of our method.

Imbalanced classification deals with learning from data with a disproportional number of samples in its classes. Traditional classifiers exhibit poor behavior when facing this kind of data because they do not take into account the imbalanced class distribution. Four main kinds of solutions exist to solve this problem: modifying the data distribution, modifying the learning algorithm for considering the imbalance representation, including the use of costs for data samples, and ensemble methods. In this paper, we adopt the second type of solution and introduce a classification algorithm for imbalanced data that uses fuzzy rough set theory and ordered weighted average aggregation. The proposal considers different strategies to build a weight vector to take into account data imbalance. Our methods are validated by an extensive experimental study, showing statistically better results than 13 other state-of-the-art methods.

The current financial crisis has stressed the need to obtain more accurate prediction models in order to decrease risk when investing money on economic opportunities. In addition, the transparency of the process followed to make the decisions in financial applications is becoming an important issue. Furthermore, there is a need to handle real-world imbalanced financial datasets without using sampling techniques that might introduce noise in the used data. In this paper, we present a compact evolutionary interval-valued fuzzy rule-based classification system, which is based on interval-valued fuzzy rule-based classification system with tuning and rule selection (IVTURSFA RC-HD) for the modeling and prediction of real-world financial applications. This proposed system allows obtaining good prediction accuracies using a small set of short fuzzy rules implying a high degree of interpretability of the generated linguistic model. Furthermore, the proposed system deals with the financial imbalanced datasets with no need for any preprocessing or sampling method and, thus, avoiding the accidental introduction of noise in the data used in the learning process. The system is also provided with a mechanism to handle examples that are not covered by any fuzzy rule in the generated rule base. To test the quality of our proposal, we will present an experimental study including 11 real-world financial datasets. We will show that the proposed system outperforms the original C4.5 decision tree, type-1, and interval-valued fuzzy counterparts that use the synthetic minority oversampling technique (SMOTE) to preprocess data and the original FURIA, which is a fuzzy approximative classifier. Furthermore, the proposed method enhances the results achieved by the cost-sensitive C4.5, and it gives competitive results when compared with FURIA using SMOTE, while our proposal avoids preprocessing techniques, and it provides interpretable models that allow obtaining more accurate results.

In this paper, we will present a wider view on the relationship between interval-valued fuzzy sets and interval type-2 fuzzy sets, where we will show that interval-valued fuzzy sets are a particular case of the interval type-2 fuzzy sets. For this reason, both concepts should be treated in a different way. In addition, the view presented in this paper will allow a more general perspective of interval type-2 fuzzy sets, which will allow representing concepts that could not be presented by interval-valued fuzzy sets.

Evolutionary prototype reduction techniques are data preprocessing methods originally developed to enhance the nearest neighbor rule. They reduce the training data by selecting or generating representative examples of a given problem. These algorithms have been designed and widely analyzed in standard classification providing very competitive results. However, its application scope can be extended to many other specific domains, such as one-class classification, in which its way of working is very interesting in order to reduce computational complexity and sensitivity to noisy data. In this contribution, we perform a first study on the usefulness of evolutionary prototype reduction methods for one-class classification. To do so, we will focus on two recent evolutionary approaches that follow very different strategies: selection and generation of examples from the training data. Both alternatives provide a resulting preprocessed data set that will be used later by a nearest neighbor one-class classifier as its training data. The results achieved support that these data reduction techniques are suitable tools to improve the performance of the nearest neighbor one-class classification.

Memetic algorithms with an appropriate trade-off between the exploration and exploitation can obtain very good results in continuous optimisation. That implies the evolutionary algorithm component should be focused in exploring the search space while the local search method exploits the achieved solutions. In a previous work, it was proposed a region-based algorithm, RMA-LSCh-CMA, adding to algorithm MA-LSCh-CMA a niching strategy that divides the domain search in equal hypercubes. The experimental results obtained, with the benchmark proposed in the CEC'2014 Special Session on RealParameter Single Objective Optimisation, show that the use of these regions allow the algorithm to obtain better results, specially in higher dimensions, and the resulting algorithm is more scalable.

Evolutionary prototype generation techniques have demonstrated their usefulness to improve the capabilities of the nearest neighbor classifier. They act as data reduction algorithms by generating representative points of a given problem. Their main purposes are to speed up the classification process and to reduce the storage requirements and sensitivity to noise of the nearest neighbor rule. Nowadays, with the increment of available data, the use of this kind of reduction techniques becomes more important. However, their applicability can be limited to problems with no more than tens of thousands of instances. In order to address this limitation, in this work we develop a two-level parallelization scheme for evolutionary prototype generation methods. Firstly, it distributes the functioning of these algorithms in several tasks based on a MapReduce framework. Then, for each one of these tasks (mappers), we accelerate the prototype generation process by using a windowing approach. This model enables evolutionary prototype generation algorithms to be applied over large-scale classification problems without accuracy loss. Our preliminary experiments using a dataset of 1 million instances show that this proposal is an appropriate tool to improve the performance of the nearest neighbor classifier with big data.

Imbalanced classification is one of the most challenging machine learning problem. Recent studies show, that often the uneven ratio of objects in classes is not the biggest factor, determining the drop of classification accuracy. It is also related to some difficulties embedded in the nature of the data. In this paper we study the different types of minority class examples and distinguish four groups of objects - safe, borderline, rare and outliers. To deal with the imbalance problem, we use a one-class classification, that is focused on a proper identification of the minority class samples. We further augment this model by incorporating the knowledge about the minority object types in the training dataset. This is done applying weighted one-class classifier and adjusting weights assigned to minority class objects, depending on their type. A strategy for calculating the new weights for minority examples is proposed. Experimental analysis, carried on a set of benchmark datasets, confirms that the proposed model can achieve a satisfactory recognition rate and often outperform other state-of-the-art methods, dedicated to the imbalanced classification.

Herrera and Martinez initiated a 2-tuple fuzzy linguistic representation model for computing with words (CWW). In addition to the Herrera and Martinez model, two different models based on linguistic 2-tuples (i.e., the model of Herrera et al. and the numerical scale model) have been developed to deal with term sets that are not uniformly and symmetrically distributed, i.e., unbalanced linguistic term sets (ULTSs). Both the model of Herrera et al. and the numerical scale model can deal with ULTSs, so a challenge is naturally proposed to analysts: how to compare these two different models. In this study, we provide a connection between the model of Herrera et al. and the numerical scale model. The results show that the model of Herrera et al. provides a new approach to set a numerical scale. Furthermore, we prove the equivalence of the linguistic computational models between the model of Herrera et al. and the numerical scale model, if the numerical scale is set based on the model of Herrera et al.

In this work we analyze the competitiveness of fuzzy rule-based systems in comparison with black box models like support vector machines to deal with fingerprint classification problems. With this aim, we carry out an experimental study applying different feature extraction models (covering almost every kind of features that are usually considered in this problem) and three fingerprint databases of different qualities. The obtained results show the good behavior of fuzzy rule-based classification systems. Observing these results, new future lines are outlined, which could improve the classification performance achieved with the current models.

Big data has become one of the emergent topics when learning from data is involved. The notorious increment in the data generation has directed the attention towards the obtaining of effective models that are able to analyze and extract knowledge from these colossal data sources. However, the vast amount of data, the variety of the sources and the need for an immediate intelligent response pose a critical challenge to traditional learning algorithms. To be able to deal with big data, we propose the usage of a linguistic fuzzy rule based classification system, which we have called Chi-FRBCS-BigData. As a fuzzy method, it is able deal with the uncertainty that is inherent to the variety and veracity of big data and because of the usage of linguistic fuzzy rules it is able to provide an interpretable and effective classification model. This method is based on the MapReduce framework, one of the most popular approaches for big data nowadays, and has been developed in two different versions: Chi-FRBCS-BigData-Max and Chi-FRBCS-BigData-Ave. The good performance of the Chi-FRBCS-BigData approach is supported by means of an experimental study over six big data problems. The results show that the proposal is able to provide competitive results, obtaining more precise but slower models in the Chi-FRBCS-BigData-Ave alternative and faster but less accurate classification results for Chi-FRBCS-BigData-Max.

Learning from data is a process to construct a model according to available training data so that it can be used to make predictions for new data. Nowadays, several software libraries are available to carry out this task, frbs is an R package which is aimed to construct models from data based on fuzzy rule based systems (FRBSs) by employing learning procedures from Computational Intelligence (e.g., neural networks and genetic algorithms) to tackle classification and regression problems. For the learning process, frbs considers well-known methods, such as Wang and Mendel's technique, ANFIS, Hy-FIS, DENFIS, subtractive clustering, SLAVE, and several others. Many options are available to perform conjunction, disjunction, and implication operators, defuzzification methods, and membership functions (e.g., triangle, trapezoid, Gaussian, etc). It has been developed in the R language which is an open-source analysis environment for scientific computing. In this paper, we also provide some examples on the usage of the package and a comparison with other software libraries implementing FRBSs. We conclude that frbs should be considered as an alternative software library for learning from data.

This paper proposes a method for designing Fuzzy Rule-Based Classification Systems to deal with One-Class Classification, where during the training phase we have access only to objects originating from a single class. However, the trained model must be prepared to deal with new, unseen adversarial objects, known as outliers. We use a Genetic Algorithm for learning the granularity, domains and fuzzy partitions of the model and we propose an ad-hoc rule generation method specific for One-Class Classification. Several datasets from UCI repository, previously transformed to one-class problems, are used in the experiments and we compare with two of the classical methods used in the One-Class community, one-class Support Vector Machines and Support Vector Data Description. Our proposal of fuzzy model obtains similar results than the other methods but presents a high interpretability due its reduced number of rules.

The necessity of dealing with uncertainty in real world problems has been a long-term research challenge that has originated different methodologies and theories. Fuzzy sets along with their extensions, such as type-2 fuzzy sets, interval-valued fuzzy sets, and Atanassov's intuitionistic fuzzy sets, have provided a wide range of tools that are able to deal with uncertainty in different types of problems. Recently, a new extension of fuzzy sets so-called hesitant fuzzy sets has been introduced to deal with hesitant situations, which were not well managed by the previous tools. Hesitant fuzzy sets have attracted very quickly the attention of many researchers that have proposed diverse extensions, several types of operators to compute with such types of information, and eventually some applications have been developed. Because of such a growth, this paper presents an overview on hesitant fuzzy sets with the aim of providing a clear perspective on the different concepts, tools and trends related to this extension of fuzzy sets.

Most of the algorithms for mining quantitative association rules focus on positive dependencies without paying particular attention to negative dependencies. The latter may be worth taking into account, however, as they relate the presence of certain items to the absence of others. The algorithms used to extract such rules usually consider only one evaluation criterion in measuring the quality of generated rules. Recently, some researchers have framed the process of extracting association rules as a multiobjective problem, allowing us to jointly optimize several measures that can present different degrees of trade-off depending on the dataset used. In this paper, we propose MOPNAR, a new multiobjective evolutionary algorithm, in order to mine a reduced set of positive and negative quantitative association rules with low computational cost. To accomplish this, our proposal extends a recent multiobjective evolutionary algorithm based on decomposition to perform an evolutionary learning of the intervals of the attributes and a condition selection for each rule, while introducing an external population and a restarting process to store all the nondominated rules found and to improve the diversity of the rule set obtained. Moreover, this proposal maximizes three objectives-comprehensibility, interestingness, and performance-in order to obtain rules that are interesting, easy to understand, and provide good coverage of the dataset. The effectiveness of the proposed approach is validated over several real-world datasets.

Consensus reaching processes in group decision making attempt to reach a mutual agreement among a group of decision makers before making a common decision. Different consensus models have been proposed by different authors in the literature to facilitate consensus reaching processes. Classical models focus on solving group decision making problems where few decision makers participate. However, nowadays, societal and technological trends that demand the management of larger scales of decision makers, such as e-democracy and social networks, add a new requirement to the solution of consensus-based group decision making problems. Dealing with such large groups implies the need for mechanisms to detect decision makers' noncooperative behaviors in consensus, which might bias the consensus reaching process. This paper presents a consensus model suitable to manage large scales of decision makers, which incorporates a fuzzy clustering-based scheme to detect and manage individual and subgroup noncooperative behaviors. The model is complemented with a visual analysis tool of the overall consensus reaching process based on self-organizing maps, which facilitates the monitoring of the process performance across the time. The consensus model presented is aimed to the solution of consensus processes involving large groups.

Fingerprints are the biometric features most used for identification. They can be characterized through some particular elements called minutiae. The identification of a given fingerprint requires the matching of its minutiae against the minutiae of other fingerprints. Hence, fingerprint matching is a key process. The efficiency of current matching algorithms does not allow their use in large fingerprint databases; to apply them, a breakthrough in running performance is necessary. Nowadays, the minutia cylinder-code (MCC) is the best performing algorithm in terms of accuracy. However, a weak point of this algorithm is its computational requirements. In this paper, we present a GPU fingerprint matching system based on MCC. The many-core computing framework provided by CUDA on NVIDIA Tesla and GeForce hardware platforms offers an opportunity to enhance fingerprint matching. Through a thorough and careful data structure, computation and memory transfer design, we have developed a system that keeps its accuracy and reaches a speed-up up to 100.8× compared with a reference sequential CPU implementation. A rigorous empirical study over captured and synthetic fingerprint databases shows the efficiency of our proposal. These results open up a whole new field of possibilities for reliable real time fingerprint identification in large databases.

Multilabel classification (MLC) has generated considerable research interest in recent years, as a technique that can be applied to many real-world scenarios. To process them with binary or multiclass classifiers, methods for transforming multilabel data sets (MLDs) have been proposed, as well as adapted algorithms able to work with this type of data sets. However, until now, few studies have addressed the problem of how to deal with MLDs having a large number of labels. This characteristic can be defined as high dimensionality in the label space (output attributes), in contrast to the traditional high dimensionality problem, which is usually focused on the feature space (by means of feature selection) or sample space (by means of instance selection). The purpose of this paper is to analyze dimensionality in the label space in MLDs, and to present a transformation methodology based on the use of association rules to discover label dependencies. These dependencies are used to reduce the label space, to ease the work of any MLC algorithm, and to infer the deleted labels in a final postprocessing stage. The proposed process is validated in an extensive experimentation with several MLDs and classification algorithms, resulting in a statistically significant improvement of performance in some cases, as will be shown.

In this paper, we present a memetic algorithm which combines in a local search chaining framework, a steady-state genetic algorithm as evolutionary algorithm and a CMA-ES as local search method. It is an extension of an already presented algorithm which uses a region-based niching strategy and which has proven to be very efficient on real parameter optimisation problems. In this new version, we propose to dynamically update the niche size in order to make it less dependent to such critical parameter. In addition, we used an automatic configuration tool to optimise its parameters, and show that the optimised version of this algorithm is significantly better than with its default parameters. We tested this algorithm on the Special Session and Competition on Real-Parameter Optimization of the IEEE Congress on Evolutionary 2013 benchmark.

Many real-world problems have several optima, and the aim of niching optimisation algorithms is to obtain the different global optima, and not only the best solution. One common technique to create niches is the clearing method that removes solutions too close to better ones. Unfortunately, clearing is very sensitive to the niche radius, and its right value depends on the problem (in real-world problems the minimum distance between optima is unknown). In this work we propose a niching algorithm that uses clearing with an adaptive niche radius, that decreases during the run. The proposal uses an external memory that stores current global optima to avoid losing found optima during the clearing process, allowing a non-elitist search. This algorithm applies this clearing method to a mesh of solutions, expanded by the generation of nodes using combination methods between the nodes, their best neighbour, and their nearest current global optima in the population (current global optima are nodes with fitness very similar to current best fitness). The proposal is tested on the competition benchmark proposed in the Special Session Niching Methods for Multimodal Optimization, and compared with other algorithms. The proposal obtains very good results detecting global optima. In comparisons with other algorithm, this proposal obtains the best results, proving to be a very competitive niching algorithm.

This paper addresses the challenging problem of fuzzy modeling in high-dimensional and large scale regression datasets. To this end, we propose a scalable two-stage method for obtaining accurate fuzzy models in high-dimensional regression problems using approximate Takagi-Sugeno-Kang Fuzzy Rule-Based Systems. In the first stage, we propose an effective Multi-Objective Evolutionary Algorithm, based on an embedded genetic Data Base learning (involved variables, granularities and a slight lateral displacement of fuzzy partitions) together with an inductive rule base learning within the same process. The second stage is a post-processing process based on a second MOEA to perform a rule selection and a fine scatter-based tuning of the Membership Functions. Moreover, it incorporates an efficient Kalman filter to estimate the coefficients of the consequent polynomial functions in the Takagi-Sugeno-Kang rules. In both stages, we include mechanisms in order to significantly improve the accuracy of the model and to ensure a fast convergence in high-dimensional regression problems. The proposed method is compared to the classical ANFIS method and to a well-known evolutionary learning algorithm for obtaining accurate TSK systems in 8 datasets with different sizes and dimensions, obtaining better results.

The estimation of the quality of the learned models in Data Mining has been traditionally carried out by means of a k-fold partition technique. However, the “random” division of the instances over the folds may results in a problem known as covariate shift, i.e. there is a different data distribution between the training and test folds. In classification with imbalanced datasets this problem is more severe. The misclassification of minority class instances due to an incorrect learning of the real boundaries caused by a not well defined data distribution, truly affects the measures of performance in this scenario. To avoid this harmful situation, we propose the use of a specific validation technique for the partitioning of the data, known as “Distribution optimally balanced stratified cross-validation”. This methodology makes the decision of placing close-by samples on different folds, so that each partition will end up with enough representatives of every region. In this contribution, we show the goodness of this methodology using Genetic Fuzzy Systems, as they are known to be robust approaches for all types of classification problems. Specifically, we have chosen the FARC-HD algorithm, a novel technique which has shown to obtain very accurate results. From the experimental analysis, which is carried out on a wide number of imbalanced datasets, we emphasize the necessity of using a proper validation methodology for extracting well founded conclusions.

Bibliometric mapping and visualization techniques represent one of the main pillars in the field of scientometrics. Traditionally, the main methodologies employed for representing data are multidimensional scaling, principal component analysis, or correspondence analysis. In this paper we aim to present a visualization methodology known as biplot analysis for representing bibliometric and science and technology indicators. A biplot is a graphical representation of multivariate data, where the elements of a data matrix are represented according to dots and vectors associated with the rows and columns of the matrix. In this paper, we explore the possibilities of applying biplot analysis in the research policy area. More specifically, we first describe and introduce the reader to this methodology and secondly, we analyze its strengths and weaknesses through 3 different case studies: countries, universities, and scientific fields. For this, we use a biplot analysis known as JK ‐biplot. Finally, we compare the biplot representation with other multivariate analysis techniques. We conclude that biplot analysis could be a useful technique in scientometrics when studying multivariate data, as well as an easy‐to‐read tool for research decision makers.

Over the past few decades, fuzzy systems have been widely used in several application fields, thanks to their ability to model complex systems. The design of fuzzy systems has been successfully performed by applying evolutionary and, in particular, genetic algorithms, and recently, this approach has been extended by using multiobjective evolutionary algorithms, which can consider multiple conflicting objectives, instead of a single one. The hybridization between multiobjective evolutionary algorithms and fuzzy systems is currently known as multiobjective evolutionary fuzzy systems. This paper presents an overview of multiobjective evolutionary fuzzy systems, describing the main contributions on this field and providing a two-level taxonomy of the existing proposals, in order to outline a well-established framework that could help researchers who work on significant further developments. Finally, some considerations of recent trends and potential research directions are presented.

Interval-valued fuzzy sets have been shown to be a useful tool to deal with the ignorance related to the definition of the linguistic labels. Specifically, they have been successfully applied to solve classification problems, performing simple modifications on the fuzzy reasoning method to work with this representation and making the classification based on a single number. In this paper, we present IVTURS, which is a new linguistic fuzzy rule-based classification method based on a new completely interval-valued fuzzy reasoning method. This inference process uses interval-valued restricted equivalence functions to increase the relevance of the rules in which the equivalence of the interval membership degrees of the patterns and the ideal membership degrees is greater, which is a desirable behavior. Furthermore, their parametrized construction allows the computation of the optimal function for each variable to be performed, which could involve a potential improvement in the system's behavior. Additionally, we combine this tuning of the equivalence with rule selection in order to decrease the complexity of the system. In this paper, we name our method IVTURS-FARC, since we use the FARC-HD method to accomplish the fuzzy rule learning process. The experimental study is developed in three steps in order to ascertain the quality of our new proposal. First, we determine both the essential role that interval-valued fuzzy sets play in the method and the need for the rule selection process. Next, we show the improvements achieved by IVTURS-FARC with respect to the tuning of the degree of ignorance when it is applied in both an isolated way and when combined with the tuning of the equivalence. Finally, the significance of IVTURS-FARC is further depicted by means of a comparison by which it is proved to outperform the results of FARC-HD and FURIA, which are two high performing fuzzy classification algorithms.

Discretization is an essential preprocessing technique used in many knowledge discovery and data mining tasks. Its main goal is to transform a set of continuous attributes into discrete ones, by associating categorical values to intervals and thus transforming quantitative data into qualitative data. In this manner, symbolic data mining algorithms can be applied over continuous data and the representation of information is simplified, making it more concise and specific. The literature provides numerous proposals of discretization and some attempts to categorize them into a taxonomy can be found. However, in previous papers, there is a lack of consensus in the definition of the properties and no formal categorization has been established yet, which may be confusing for practitioners. Furthermore, only a small set of discretizers have been widely considered, while many other methods have gone unnoticed. With the intention of alleviating these problems, this paper provides a survey of discretization methods proposed in the literature from a theoretical and empirical perspective. From the theoretical perspective, we develop a taxonomy based on the main properties pointed out in previous research, unifying the notation and including all the known methods up to date. Empirically, we conduct an experimental study in supervised classification involving the most representative and newest discretizers, different types of classifiers, and a large number of data sets. The results of their performances measured in terms of accuracy, number of intervals, and inconsistency have been verified by means of nonparametric statistical tests. Additionally, a set of discretizers are highlighted as the best performing ones.

Memetic algorithms with an appropriate trade-off between the exploration and exploitation can obtain very good results in continuous optimisation. That implies the evolutionary algorithm component should be focused in exploring the search space while the local search method exploits the achieved solutions. In a previous work, it was proposed a memetic algorithm, MA-LSCh-CMA, that was able to work with a local search method, CMA-ES, with a great exploitation factor, but without a mechanism to maintain diversity and avoid competition between the evolutionary algorithm and CMA-ES. In this work, we propose a variation of this algorithm, called RMA-LSCh-CMA, adding a niching strategy that divide the domain search in equal hypercubes. The experimental results obtained show that the new version is statistically better than the previous one and is very competitive in comparisons with the state-of-the-art algorithm IPOP-CMA-ES, obtaining equivalent results on the medium and higher dimensions, although slightly better in the higher dimension.

Population-based Meta-heuristics are algorithms that can obtain very good results for complex continuous optimisation problems, using the information of a population of solutions. In these algorithms the distribution of solutions is crucial because it has a strong influence of the exploration new regions. In this work, we present a population algorithm, Variable Mesh Optimisation (VMO), in which a set of nodes (potential solutions) is distributed as a mesh. This mesh is initially homogeneously distributed, and then the mesh evolves to a heterogeneous structure resampling the space toward the best neighbours, maintaining at the same time a controlled diversity (avoiding solutions too close to each other). We use a benchmark of multimodal continuous functions to study the influence of the different components of the proposal, and to compare the proposed algorithm with other basic population-based metaheuristics in the literature. The results show that VMO is a very competitive algorithm.

When considering data sets characterized by a large number of instances, the computational time required to apply Genetic Algorithms for generating Fuzzy Rule-Based Classifiers increases considerably, mainly due to the fitness evaluation. Another important problem associated to these kinds of data sets is an undesired increase of the obtained model complexity.

This paper proposes a Genetic Algorithm for jointly performing a feature selection and granularity learning for Fuzzy Rule-Based Classification Systems in the scenario of highly imbalanced data-sets. We refer to imbalanced data-sets when the class distribution is not uniform, a situation that it is present in many real application areas. The aim of this work is to get more compact models by selecting the adequate variables and adapting the number of fuzzy labels for each problem, improving the interpretability of the model. The experimental analysis is carried out over a wide range of highly imbalanced data-sets and uses the statistical tests suggested in the specialized literature.

This article presents a new open‐source software tool, SciMAT , which performs science mapping analysis within a longitudinal framework. It provides different modules that help the analyst to carry out all the steps of the science mapping workflow. In addition, SciMAT presents three key features that are remarkable in respect to other science mapping software tools: (a) a powerful preprocessing module to clean the raw bibliographical data, (b) the use of bibliometric measures to study the impact of each studied element, and (c) a wizard to configure the analysis.

The nearest neighbor classifier is one of the most used and well-known techniques for performing recognition tasks. It has also demonstrated itself to be one of the most useful algorithms in data mining in spite of its simplicity. However, the nearest neighbor classifier suffers from several drawbacks such as high storage requirements, low efficiency in classification response, and low noise tolerance. These weaknesses have been the subject of study for many researchers and many solutions have been proposed. Among them, one of the most promising solutions consists of reducing the data used for establishing a classification rule (training data) by means of selecting relevant prototypes. Many prototype selection methods exist in the literature and the research in this area is still advancing. Different properties could be observed in the definition of them, but no formal categorization has been established yet. This paper provides a survey of the prototype selection methods proposed in the literature from a theoretical and empirical point of view. Considering a theoretical point of view, we propose a taxonomy based on the main characteristics presented in prototype selection and we analyze their advantages and drawbacks. Empirically, we conduct an experimental study involving different sizes of data sets for measuring their performance in terms of accuracy, reduction capabilities, and runtime. The results obtained by all the methods studied have been verified by nonparametric statistical tests. Several remarks, guidelines, and recommendations are made for the use of prototype selection for nearest neighbor classification.

Dealing with uncertainty is always a challenging problem, and different tools have been proposed to deal with it. Recently, a new model that is based on hesitant fuzzy sets has been presented to manage situations in which experts hesitate between several values to assess an indicator, alternative, variable, etc. Hesitant fuzzy sets suit the modeling of quantitative settings; however, similar situations may occur in qualitative settings so that experts think of several possible linguistic values or richer expressions than a single term for an indicator, alternative, variable, etc. In this paper, the concept of a hesitant fuzzy linguistic term set is introduced to provide a linguistic and computational basis to increase the richness of linguistic elicitation based on the fuzzy linguistic approach and the use of context-free grammars by using comparative terms. Then, a multicriteria linguistic decision-making model is presented in which experts provide their assessments by eliciting linguistic expressions. This decision model manages such linguistic expressions by means of its representation using hesitant fuzzy linguistic term sets.

In this paper, we propose new aggregation functions for the pairwise comparison of alternatives in fuzzy preference modeling. More specifically, we introduce the concept of a grouping function, i.e., a specific type of aggregation function that combines two degrees of support (weak preference) into a degree of information or, say, a degree of comparability between two alternatives, and we relate this new concept to that of incomparability. Grouping functions of this type complement the existing concept of overlap functions in a natural way, since the latter can be used to turn two degrees of weak preference into a degree of indifference. We also define the so-called generalized bientropic functions that allow for a unified representation of overlap and grouping functions. Apart from analyzing mathematical properties of these types of functions and exploring relationships between them, we elaborate on their use in fuzzy preference modeling and decision making. We present an algorithm to elaborate on an alternative preference ranking that penalizes those alternatives for which the expert is not sure of his/her preference.

In this paper, we extend the study of the intelligent transportation system (ITS) topic evolution presented by Li et al. To do so, we apply an approach that combines both H-index-based performance analysis and science mapping to detect, visualize, and evaluate conceptual ITS themes and ITS thematic areas published by the journal IEEE Transactions on Intelligent Transport Systems during the decade (2000–2009). The primary consequence of this is the detection of three important thematic areas: COMPUTER-VISION and TRAFFIC-FLOW, which are related to research in ITS applied to vehicles, and AIRCRAFT-TRAFFIC, which is related to research in ITS applied to aircraft/airport.

Cross-validation is a very commonly employed technique used to evaluate classifier performance. However, it can potentially introduce dataset shift, a harmful factor that is often not taken into account and can result in inaccurate performance estimation. This paper analyzes the prevalence and impact of partition-induced covariate shift on different k-fold cross-validation schemes. From the experimental results obtained, we conclude that the degree of partition-induced covariate shift depends on the cross-validation scheme considered. In this way, worse schemes may harm the correctness of a single-classifier performance estimation and also increase the needed number of repetitions of cross-validation to reach a stable performance estimation.

Cooperative coevolution is a successful trend of evolutionary computation which allows us to define partitions of the domain of a given problem, or to integrate several related techniques into one, by the use of evolutionary algorithms. It is possible to apply it to the development of advanced classification methods, which integrate several machine learning techniques into a single proposal. A novel approach integrating instance selection, instance weighting, and feature weighting into the framework of a coevolutionary model is presented in this paper. We compare it with a wide range of evolutionary and nonevolutionary related methods, in order to show the benefits of the employment of coevolution to apply the techniques considered simultaneously. The results obtained, contrasted through nonparametric statistical tests, show that our proposal outperforms other methods in the comparison, thus becoming a suitable tool in the task of enhancing the nearest neighbor classifier.

Classifier learning with data-sets that suffer from imbalanced class distributions is a challenging problem in data mining community. This issue occurs when the number of examples that represent one class is much lower than the ones of the other classes. Its presence in many real-world applications has brought along a growth of attention from researchers. In machine learning, the ensemble of classifiers are known to increase the accuracy of single classifiers by combining several of them, but neither of these learning techniques alone solve the class imbalance problem, to deal with this issue the ensemble learning algorithms have to be designed specifically. In this paper, our aim is to review the state of the art on ensemble techniques in the framework of imbalanced data-sets, with focus on two-class problems. We propose a taxonomy for ensemble-based methods to address the class imbalance where each proposal can be categorized depending on the inner ensemble methodology in which it is based. In addition, we develop a thorough empirical comparison by the consideration of the most significant published approaches, within the families of the taxonomy proposed, to show whether any of them makes a difference. This comparison has shown the good behavior of the simplest approaches which combine random undersampling techniques with bagging or boosting ensembles. In addition, the positive synergy between sampling techniques and bagging has stood out. Furthermore, our results show empirically that ensemble-based algorithms are worthwhile since they outperform the mere use of preprocessing techniques before learning the classifier, therefore justifying the increase of complexity by means of a significant enhancement of the results.

The nearest neighbor (NN) rule is one of the most successfully used techniques to resolve classification and pattern recognition tasks. Despite its high classification accuracy, this rule suffers from several shortcomings in time response, noise sensitivity, and high storage requirements. These weaknesses have been tackled by many different approaches, including a good and well-known solution that we can find in the literature, which consists of the reduction of the data used for the classification rule (training data). Prototype reduction techniques can be divided into two different approaches, which are known as prototype selection and prototype generation (PG) or abstraction. The former process consists of choosing a subset of the original training data, whereas PG builds new artificial prototypes to increase the accuracy of the NN classification. In this paper, we provide a survey of PG methods specifically designed for the NN rule. From a theoretical point of view, we propose a taxonomy based on the main characteristics presented in them. Furthermore, from an empirical point of view, we conduct a wide experimental study that involves small and large datasets to measure their performance in terms of accuracy and reduction capabilities. The results are contrasted through nonparametrical statistical tests. Several remarks are made to understand which PG models are appropriate for application to different datasets.

Musical Genre is part of the basic information required for classifying musical audio, and fundamental for music information retrieval systems. The problem of automatic musical genre detection has attracted large attention in the last decade, due to the emergence of digital music databases and Internet. Although a number of techniques has been applied to the problem, no general solution still exists, due to the imprecise features that properly define musical genre. This paper presents a preliminary attempt to apply Fuzzy Rule-Based System (FRBS) in cooperation with Evolutionary Algorithms to musical genre classification. The novelty of the approach -which allows us to use fuzzy information extracted from audio filesis aligned with the fuzzy nature of the problem at hand, where no clear-cut rules are available for the classification. Preliminary results presented allows to foresee the potential of the technique.

Recent studies in smart homes have proposed methods to use a laser pointer for interacting with home devices, which represents a more user-friendly and less expensive home device control environment. However, detecting the laser spot on the original non-filtered images, using standard and non expensive cameras, and considering real home environments with varying conditions, is currently an open problem. In this paper we propose a hybrid technique, combining a classic technique used in image detection processes, such as Template Matching, with an evolutionary learning of a Fuzzy Rule Based Systems for the laser spot detection system in real home environments. This proposal improves the success rate in images without laser spot of the previous classical and non-classical algorithms used for detecting the laser spot in previous works, decreasing the detection of the false offs which could lead to dangerous situations. Experimental results on a real home environment show the effectiveness of the proposed approach.

Fuzzy decision trees are widely employed to face classification problems since they combine the high interpretability given by the decision tree and the capability of management of the uncertainty inherent to fuzzy logic. However, the success of fuzzy systems in general depends, to a large degree, on the choice of the membership functions. For this reason, we propose to model the linguistic labels by means of Interval-Valued Fuzzy Sets to take into account the ignorance related to their definition. On the other hand, we define an evolutionary method to tune the shape of the Interval-Valued Fuzzy Sets looking for the best ignorance degree that each Interval-Valued Fuzzy Set represents. In this contribution, we will make use of the fuzzy ID3 algorithm as a base technique from which to apply our methodology. The experimental study shows how our methodology enhances the performance of the base fuzzy decision tree. Furthermore, we compare our approach with respect to four state-of-the-art fuzzy decision trees and C4.5 as a representative algorithm for crisp decision trees. The goodness of our proposal is tested on a large collection of data-sets and it is supported by an exhaustive statistical analysis.

This paper studies the behavior of a multiobjective Genetic Algorithm for jointly performing a feature selection and granularity learning for Fuzzy Rule-Based Classification Systems in the scenario of imbalanced data-sets. We refer to imbalanced data-sets when the class distribution is not uniform, a situation that it is present in many real application areas. We consider two different measures, one for the precision of the model and other for its complexity as the two objectives to optimize. In one previous approach, we aggregate these two measures in a single-objective Genetic Algorithm, and thus, a multiobjective approach of that Genetic Algorithm would yield a set of models with different trade-off between high accuracy and low complexity rather than a unique model, provided by the single-objective Genetic Algorithm. The experimental analysis, carried out over a wide range of imbalanced data-sets, shows that our approach is able to obtain a set of models with good trade-off between the two objectives considered but it is an open problem how to select the solution with best prediction ability from the whole set of solutions obtained.

Data mining is most commonly used in attempts to induce association rules from database. Recently, some researchers have suggested the extraction of association rules as a multi-objective problem, removing some of the limitations of current approaches. In this way, we can jointly optimize quality measures which can present different degrees of tradeoff depending on the database used and the type of information can be extracted from it. In this work, we extend the well-known multi-objective evolutionary algorithms NSGA-II to perform an evolutionary learning of the intervals of attributes and a condition selection in order to mine a set of quantitative association rules with a good trade-off between interpretability and accuracy. To do that, this method considers three objectives, maximize the interestingness, comprehensibility and performance. Moreover, this method follows a database-independent approach which does not rely upon minimum support and minimum confidence thresholds. The results obtained over two real-world databases demonstrate the effectiveness of the proposed approach.

The presence of noise is common in any real-world dataset and may adversely affect the accuracy, construction time and complexity of the classifiers in this context. Traditionally, many algorithms have incorporated mechanisms to deal with noisy problems and reduce noise's effects on performance; they are called robust learners. The C4.5 crisp algorithm is a well-known example of this group of methods. On the other hand, models built by Fuzzy Rule Based Classification Systems are widely recognized for their robustness to imperfect data, but also for their interpretability. The aim of this contribution is to analyze the good behavior and robustness of Fuzzy Rule Based Classification Systems when noise is present in the examples' class labels, especially versus robust learners. In order to accomplish this study, a large number of datasets are created by introducing different levels of noise into the class labels in the training sets. We compare a Fuzzy Rule Based Classification System, the Fuzzy Unordered Rule Induction Algorithm, with respect to the C4.5 classic robust learner which is considered tolerant to noise. From the results obtained it is possible to observe that Fuzzy Rule Based Classification Systems have a good tolerance, in comparison to the C4.5 algorithm, to class noise.

Science mapping aims to build bibliometric maps that describe how specific disciplines, scientific domains, or research fields are conceptually, intellectually, and socially structured. Different techniques and software tools have been proposed to carry out science mapping analysis. The aim of this article is to review, analyze, and compare some of these software tools, taking into account aspects such as the bibliometric techniques available and the different kinds of analysis.

The inductive learning of fuzzy rule-based classification systems suffers from exponential growth of the fuzzy rule search space when the number of patterns and/or variables becomes high. This growth makes the learning process more difficult and, in most cases, it leads to problems of scalability (in terms of the time and memory consumed) and/or complexity (with respect to the number of rules obtained and the number of variables included in each rule). In this paper, we propose a fuzzy association rule-based classification method for high-dimensional problems, which is based on three stages to obtain an accurate and compact fuzzy rule-based classifier with a low computational cost. This method limits the order of the associations in the association rule extraction and considers the use of subgroup discovery, which is based on an improved weighted relative accuracy measure to preselect the most interesting rules before a genetic postprocessing process for rule selection and parameter tuning. The results that are obtained more than 26 real-world datasets of different sizes and with different numbers of variables demonstrate the effectiveness of the proposed approach.

Linguistic fuzzy modeling in high-dimensional regression problems poses the challenge of exponential-rule explosion when the number of variables and/or instances becomes high. One way to address this problem is by determining the used variables, the linguistic partitioning and the rule set together, in order to only evolve very simple, but still accurate models. However, evolving these components together is a difficult task, which involves a complex search space. In this study, we propose an effective multiobjective evolutionary algorithm that, based on embedded genetic database (DB) learning (involved variables, granularities, and slight fuzzy-partition displacements), allows the fast learning of simple and quite-accurate linguistic models. Some efficient mechanisms have been designed to ensure a very fast, but not premature, convergence in problems with a high number of variables. Further, since additional problems could arise for datasets with a large number of instances, we also propose a general mechanism for the estimation of the model error when using evolutionary algorithms, by only considering a reduced subset of the examples. By doing so, we can also apply a fast postprocessing stage for further refining the learned solutions. We tested our approach on 17 real-world datasets with different numbers of variables and instances. Three well-known methods based on embedded genetic DB learning have been executed as references. We compared the different approaches by applying nonparametric statistical tests for multiple comparisons. The results confirm the effectiveness of the proposed method not only in terms of scalability but in terms of the simplicity and generalizability of the obtained models as well.

Memetic algorithms are effective algorithms to obtain reliable and accurate solutions for complex continuous optimization problems. Nowadays, high dimensional optimization problems are an interesting field of research. The high dimensionality introduces new problems for the optimization process, requiring more scalable algorithms that, at the same time, could explore better the higher domain space around each solution. In this work, we proposed a memetic algorithm, MA-SW-Chains, for large scale global optimization. This algorithm assigns to each individual a local search intensity that depends on its features, by chaining different local search applications. MA-SW-Chains is an adaptation to large scale optimization of a previous algorithm, MA-CMA-Chains, to improve its performance on high-dimensional problems. Finally, we present the results obtained by our proposal using the benchmark problems defined in the Special Session of Large Scale Global Optimization on the IEEE Congress on Evolutionary Computation in 2010.

Nearest neighbor is one of the most successfully used techniques for performing classification and pattern recognition tasks. Its simplicity and effectiveness justify the use of this technique in certain domains but it however presents several drawbacks referring to time response, noise sensitivity and storage requirements. Several solutions have been proposed in order to alleviate these problems, such as improving the technique for speeding up or carrying out a data reduction process. Prototype generation is a suitable process for data reduction that allows to fit a data set for nearest neighbor classification. Position adjustment of prototypes is a successful technique within the prototype generation methodology. Evolutionary algorithms are adaptive methods based on natural evolution that may be used for search and optimization. Position adjustment of prototypes can be viewed as a search problem, thus it could be solved using evolutionary algorithms. In this paper, we perform a preliminary study on the use of differential evolution algorithms to the prototype generation problem. Differential evolution models are compared with other algorithms for adjusting the position of prototypes and the results are contrasted through non-parametrical statistical tests. The results show that some differential evolution models consistently outperform previously proposed methods.

In this paper we present a new approach for laser-based environment device control systems by laser pointer for handicapped people. The paper proposes the design of a Fuzzy Rule Base System for laser pointer detection. The idea is to improve the success rate of the previous approaches decreasing as much as possible the false offs, i.e., the detection of a false laser spot (since this could lead to dangerous situations). To this end, Genetic Fuzzy Systems have also been employed for improving the laser spot system detection thus reducing the system false offs, that is the main objective in this problem. The system presented in this paper, using a Fuzzy Rule Base System adjusted by a Genetic Algorithm, shows a better success rate, and the most important thing, the not desired false offs are completely avoided.

When dealing with problems using Fuzzy Rule Based Classification Systems it is difficult to know in advance whether the model will perform well or badly. In this work we present an automatic extraction method to determine the domains of competence of Fuzzy Rule Based Classification Systems As a case of study we use the Fuzzy Hybrid Genetic Based Machine Learning method. We consider twelve metrics of data complexity in order to analyze the behavior patterns of this method, obtaining intervals of such data complexity measures with good or bad performance of it. Combining these intervals we obtain rules that describe both good or bad behaviors of the Fuzzy Rule Based Classification System mentioned. These rules allow describe both good or bad behaviors of the Fuzzy Rule Based Classification Systems mentioned, allowing us to characterize the response quality of the methods from the data set complexity metrics of a given data set. Thus, we can establish the domains of competence of the Fuzzy Rule Based Classification Systems considered, making it possible to establish when the method will perform well or badly prior to its application.

Fuzzy Rule-Based Classification Systems are a widely used tool in Data Mining because of the interpretability given by the concept of linguistic label. However, the use of this type of models implies a degree of uncertainty in the definition of the fuzzy partitions. In this work we will use the concept of Interval-Valued Fuzzy Set to deal with this problem. The aim of this contribution is to show the improvement in the performance of linguistic Fuzzy Rule-Based Classification Systems afterward the application of a cooperative tuning methodology between the tuning of the amplitude of the support and the lateral tuning (based on the 2-tuples fuzzy linguistic model) applied to the linguistic labels modeled with Interval-Valued Fuzzy Sets.

Classification in imbalanced domains has become one of the most relevant problems within the area of Machine Learning at the present. This problem has raised in significance due to its presence in many real applications and it occurs when the distribution of the available examples to carry out the learning process is very different between the classes (often for binary class data-sets). Usually, the underrepresented class is the concept of the most interest for the problem, being the cost derived from a misclassification of these examples much higher than that of the remaining examples. In this work we analyze the behaviour of a cost-sensitive learning method for Fuzzy Rule Based Classification Systems in the scenario of high imbalanced data-sets. Specifically, we focus on one representative rule learning approach for Genetic Fuzzy Systems, the Fuzzy Hybrid Genetics-Based Machine Learning algorithm. The experimental results show how our cost-sensitive approach in this type of domains will help us to obtain very accurate solutions in shorter training times and also with a lower complexity with respect to other possibilities proposed for classification with imbalanced problems such as the use of preprocessing to rebalance the class distribution.

The classification of imbalanced data is a well-studied topic in data mining. However, there is still a lack of understanding of the factors that make the problem difficult. In this work, we study the two main reasons that make the classification of imbalanced datasets complex: overlapping and data fracture. We present a Genetic Programming-based feature extraction method driven by Rough Set Theory to help visualize the data in a bidimensional graph, to better understand how the presence of overlapping and data fractures affect classification performance.

Memetic algorithms with continuous local search methods have arisen as effective tools to address the difficulty of obtaining reliable solutions of high precision for complex continuous optimisation problems. There exists a group of continuous local search algorithms that stand out as exceptional local search optimisers. However, on some occasions, they may become very expensive, because of the way they exploit local information to guide the search process. In this paper, they are called intensive continuous local search methods. Given the potential of this type of local optimisation methods, it is interesting to build prospective memetic algorithm models with them. This paper presents the concept of local search chain as a springboard to design memetic algorithm approaches that can effectively use intense continuous local search methods as local search operators. Local search chain concerns the idea that, at one stage, the local search operator may continue the operation of a previous invocation, starting from the final configuration (initial solution, strategy parameter values, internal variables, etc.) reached by this one. The proposed memetic algorithm favours the formation of local search chains during the memetic algorithm run with the aim of concentrating local tuning in search regions showing promise. In order to study the performance of the new memetic algorithm model, an instance is implemented with CMA-ES as an intense local search method. The benefits of the proposal in comparison to other kinds of memetic algorithms and evolutionary algorithms proposed in the literature to deal with continuous optimisation problems are experimentally shown. Concretely, the empirical study reveals a clear superiority when tackling high-dimensional problems.

DataMining is most commonly used in attempts to induce association rules from transaction data which can help decision-makers easily analyze the data and make good decisions regarding the domains concerned. Most conventional studies are focused on binary or discrete-valued transaction data, however the data in real-world applications usually consists of quantitative values. In the last years, many researches have proposed Genetic Algorithms for mining interesting association rules from quantitative data. In this paper, we present a study of three genetic association rules extraction methods to show their effectiveness for mining quantitative association rules. Experimental results over two real-world databases are showed.

Abstract Thomson Reuters' Web of Science (WoS) is undoubtedly a great tool for scientiometrics purposes. It allows one to retrieve and compute different measures such as the total number of papers that satisfy a particular condition; however, it also is well known that this tool imposes several different restrictions that make obtaining certain results difficult. One of those constraints is that the tool does not offer the total count of documents in a dataset if it is larger than 100,000 items. In this article, we propose and analyze different approaches that involve partitioning the search space (using the Source field) to retrieve item counts for very large datasets from the WoS. The proposed techniques improve previous approaches: They do not need any extra information about the retrieved dataset (thus allowing completely automatic procedures to retrieve the results), they are designed to avoid many of the restrictions imposed by the WoS, and they can be easily applied to almost any query. Finally, a description of WoS Query Partitioner, a freely available and online interactive tool that implements those techniques, is presented.

The classification problem can be addressed by numerous techniques and algorithms which belong to different paradigms of machine learning. In this paper, we are interested in evolutionary algorithms, the so-called genetics-based machine learning algorithms. In particular, we will focus on evolutionary approaches that evolve a set of rules, i.e., evolutionary rule-based systems, applied to classification tasks, in order to provide a state of the art in this field. This paper has a double aim: to present a taxonomy of the genetics-based machine learning approaches for rule induction, and to develop an empirical analysis both for standard classification and for classification with imbalanced data sets. We also include a comparative study of the genetics-based machine learning (GBML) methods with some classical non-evolutionary algorithms, in order to observe the suitability and high potential of the search performed by evolutionary algorithms and the behavior of the GBML algorithms in contrast to the classical approaches, in terms of classification accuracy.

A non-dominated multiobjective evolutionary algorithm for extracting fuzzy rules in subgroup discovery (NMEEF-SD) is described and analyzed in this paper. This algorithm, which is based on the hybridization between fuzzy logic and genetic algorithms, deals with subgroup-discovery problems in order to extract novel and interpretable fuzzy rules of interest, and the evolutionary fuzzy system NMEEF-SD is based on the well-known Non-dominated Sorting Genetic Algorithm II (NSGA-II) model but is oriented toward the subgroup-discovery task using specific operators to promote the extraction of interpretable and high-quality subgroup-discovery rules. The proposal includes different mechanisms to improve diversity in the population and permits the use of different combinations of quality measures in the evolutionary process. An elaborate experimental study, which was reinforced by the use of nonparametric tests, was performed to verify the validity of the proposal, and the proposal was compared with other subgroup discovery methods. The results show that NMEEF-SD obtains the best results among several algorithms studied.

In this paper, we propose an index that helps preserve the semantic interpretability of linguistic fuzzy models while a tuning of the membership functions (MFs) is performed. The proposed index is the aggregation of three metrics that preserve the original meanings of the MFs as much as possible while a tuning of their definition parameters is performed. Additionally, rule-selection mechanisms can be used to reduce the model complexity, which involves another important interpretability aspect. To this end, we propose a postprocessing multiobjective evolutionary algorithm that performs rule selection and tuning of fuzzy-rule-based systems with three objectives: accuracy, semantic interpretability maximization, and complexity minimization. We tested our approach on nine real-world regression datasets. In order to analyze the interaction between the fuzzy-rule-selection approach and the tuning approach, these are also individually proved in a multiobjective framework and compared with their respective single-objective counterparts. We compared the different approaches by applying nonparametric statistical tests for pairwise and multiple comparisons, taking into consideration three representative points from the obtained Pareto fronts in the case of the multiobjective-based approaches. Results confirm the effectiveness of our approach, and a wide range of solutions is obtained, which are not only more interpretable but are also more accurate.

Nearest prototype methods are a successful trend of many pattern classification tasks. However, they present several shortcomings such as time response, noise sensitivity, and storage requirements. Data reduction techniques are suitable to alleviate these drawbacks. Prototype generation is an appropriate process for data reduction, which allows the fitting of a dataset for nearest neighbor (NN) classification. This brief presents a methodology to learn iteratively the positioning of prototypes using real parameter optimization procedures. Concretely, we propose an iterative prototype adjustment technique based on differential evolution. The results obtained are contrasted with nonparametric statistical tests and show that our proposal consistently outperforms previously proposed methods, thus becoming a suitable tool in the task of enhancing the performance of the NN classifier.

Classification is one of the most researched questions in machine learning and data mining. A wide range of real problems have been stated as classification problems, for example credit scoring, bankruptcy prediction, medical diagnosis, pattern recognition, text categorization, software quality assessment, and many more. The use of evolutionary algorithms for training classifiers has been studied in the past few decades. Genetic programming (GP) is a flexible and powerful evolutionary technique with some features that can be very valuable and suitable for the evolution of classifiers. This paper surveys existing literature about the application of genetic programming to classification, to show the different ways in which this evolutionary algorithm can help in the construction of accurate and reliable classifiers.

Memetic algorithms arise as very effective algorithms to obtain reliable and high accurate solutions for complex continuous optimization problems. Nowadays, high dimensional optimization problems are an interesting field of research. The high dimensionality introduces new problems for the optimization process, making recommendable to test the behavior of the optimization algorithms to large-scale problems. The local search method must be applied with a higher intensity, specially to most promising solutions, to explore the higher domain space around each solution. In this work, we present a preliminar study of a memetic algorithm that assigns to each individual a local search intensity that depends on its features, by chaining different local search applications. This algorithm have obtained good results in continuous optimization and we study whether is a good algorithm for large scale optimizations problems. We make experiments of our proposal using the benchmark problems defined in the Special Session or Competition on Large Scale Global Optimisation, on the IEEE Congress on Evolutionary Computation in 2008. First, we test different local search methods to identify the best one. Then, we compare the proposed algorithm with the algorithms used into the competition, obtaining that our proposal is a very promising algorithm for this type of high-dimensional problems: with dimension 500 our proposal is the second best of the compared algorithms, and the best memetic algorithm.

Recently, multiobjective evolutionary algorithms have been applied to improve the difficult tradeoff between interpretability and accuracy of fuzzy rule-based systems. It is known that both requirements are usually contradictory, however, these kinds of algorithms can obtain a set of solutions with different trade-offs. The application of multiobjective evolutionary algorithms to fuzzy rule-based systems is often referred to as multiobjective genetic fuzzy systems. The first study on multiobjective genetic fuzzy systems was multiobjective genetic fuzzy rule selection in order to simultaneously achieve accuracy maximization and complexity minimization. This approach is based on the generation of a set of candidate fuzzy classification rules by considering a previously fixed granularity or multiple fuzzy partitions with different granularities for each attribute. Then, a multiobjective evolutionary optimization algorithm is applied to perform fuzzy rule selection. Although the multiple granularity approach is one of the most promising approaches, its interpretability loss has often been pointed out. In this work, we propose a mechanism to generate single granularity-based fuzzy classification rules for multiobjective genetic fuzzy rule selection. This mechanism is able to specify appropriate single granularities for fuzzy rule extraction before performing multiobjective genetic fuzzy rule selection. The results show that the performance of the obtained classifiers can be even improved by avoiding multiple granularities, which increases the linguistic interpretability of the obtained models.

The interpretability of the results obtained and the quality measures used both to extract and evaluate the rules are two key aspects of Subgroup Discovery. In this study, we analyse the influence of the type of rule used to extract knowledge in Subgroup Discovery, and the quality measures more adapted to the evolutionary algorithms for Subgroup Discovery developed so far. The adaptation of the NMEF-SD algorithm to extract disjunctive formal norm rules is also presented.