Abstract Tropospheric ozone forecasting is critical for public health, yet the deep learning models that achieve high accuracy often function as black boxes. This lack of transparency, along with the inability of popular explainability techniques like SHapley Additive exPlanations (SHAP) to capture essential temporal dependencies, limits their practical utility and trustworthiness in environmental management. To address this, we propose a novel framework, eXplainable Deep Learning with Spectral Co-clustering for Time Series, that integrates spectral co-clustering to enhance forecasting performance and provide post-hoc structured interpretability for air-quality time series. The methodology comprises data preprocessing, feature engineering (including lagging, rolling statistics, and time-based features), and deep learning architectures (Multilayer Perceptron, Gated Recurrent Unit, and hybrid models). Bayesian optimization is used to fine-tune hyperparameters. The core contribution is a spectral co-clustering technique that simultaneously partitions features and time instances into co-clusters, revealing critical inter-feature relationships and temporal patterns that drive predictions. The framework was rigorously validated through extensive experiments on data from five air quality monitoring stations. The proposed approach achieved RMSE values ranging from 0.73 to 6.08, significantly outperforming existing methods, including a temporal LSTM baseline, with performance improvements of approximately 59.19% to 95.65%. Results demonstrate that the proposed approach not only achieves high forecasting accuracy but also, through post-hoc heatmap visualizations of the objectively selected best-performing co-cluster, identifies the key features and time periods governing model predictions, thereby offering an interpretable understanding of the temporal and feature-level drivers associated with ozone variability. Thus, a transparent and effective solution for ozone forecasting is proposed, with a modular design generalizable to other environmental time series prediction tasks.

Effective and efficient disease detection is crucial, particularly for economically important crops like cotton.In this paper, we move from the initial development of a deep-learning model for cotton leaf disease detection, called Deep-CCNet, to a more comprehensive comparison of different feature selection algorithms, such as RainWater Algorithm, Particle Swarm Optimization, Bee Evolutionary Algorithm, Genetic Algorithm, and Binary Dragonfly Algorithm.Although Deep-CCNet achieved satisfactory classification performance, the goal of this study is to improve the classification performance and efficiency of deep learning models with meta-heuristic feature selection techniques.This study aims to determine which feature selection method achieves the best balance between performance and computational efficiency.We used the Kaggle "cotton leaf disease dataset", which has 1,711 images from four classes (namely curl virus, bacterial blight, fusarium wilt, and healthy leaf images), to compare these techniques systematically.Our research attempts to find the most effective method that maximizes model performance while minimizing computing resources, in addition to benchmarking the computational and performance parameters of each approach.The results of this study provide a new approach for the choice of feature selection methods in plant pathology, leading to better early disease diagnosis and increased crop resilience via efficient farming practices.

This work examines the combined effect of segmentation and data augmentation, two key preprocessing strategies often studied separately, on AI model classification performance and explainability. Three key experiments are conducted. First, the modified MobileNetV2 is applied to 21,165 raw images from the COVID-19 Radiography Database. While classification results are strong, Grad-CAM explanations misfocus on areas below the chest. Second, U-Net segmentation crops chest regions, and applying rotation, flipping, and brightness adjustment achieves a balance between accuracy and explainability. Third, precise cropping using U-Net segmentation masks isolates chest areas but slightly degrades classifier performance without further explainability gains. Findings suggest that combining U-Net segmentation with augmentation enhances explainability while maintaining model precision for COVID-19 detection. Their integration offers a trade-off between accuracy and explainability, reinforcing their complementary role in medical image analysis.

Hyperparameter optimization is a pivotal step in enhancing model performance within machine learning. Traditionally, this challenge is addressed through metaheuristics, which efficiently explore large search spaces to uncover optimal solutions. However, implementing these techniques can be complex without adequate development tools, which is the primary focus of this paper. Hence, we introduce MetaGen , a novel Python package designed to provide a comprehensive framework for developing and evaluating metaheuristic algorithms. MetaGen follows best practices in Python design, ensuring minimalistic code implementation, intuitive comprehension, and full flexibility in solution representation. The package defines two distinct user roles: Developers, responsible for algorithm implementation for hyperparameter optimization, and Solvers, who leverage pre-implemented metaheuristics to address optimization problems. Beyond algorithm implementation, MetaGen facilitates benchmarking through built-in test functions, ensuring standardized performance comparisons. It also provides automated reporting and visualization tools to analyze optimization progress and outcomes effectively. Furthermore, its modular design allows distribution and integration into existing machine learning workflows. Several illustrative use cases are presented to demonstrate its adaptability and efficacy. The package, along with code, a user manual, and supplementary materials, is available at: https://github.com/Data-Science-Big-Data-Research-Lab/MetaGen .

Artificial intelligence and deep learning are powerful tools for extracting knowledge from large datasets, particularly in healthcare. However, their black-box nature raises interpretability concerns, especially in high-stakes applications. Existing eXplainable Artificial Intelligence methods often focus solely on visualization or rule-based explanations, limiting interpretability's depth and clarity. This work proposes a novel explainable AI method specifically designed for medical image analysis, integrating statistical, visual, and rule-based explanations to improve transparency in deep learning models. Statistical features are derived from deep features extracted using a custom Mobilenetv2 model. A two-step feature selection method - zero-based filtering with mutual importance selection - ranks and refines these features. Decision tree and RuleFit models are employed to classify data and extract human-readable rules. Additionally, a novel statistical feature map overlay visualization generates heatmap-like representations of three key statistical measures (mean, skewness, and entropy), providing both localized and quantifiable visual explanations of model decisions. The proposed method has been validated on five medical imaging datasets - COVID-19 radiography, ultrasound breast cancer, brain tumor magnetic resonance imaging, lung and colon cancer histopathological, and glaucoma images - with results confirmed by medical experts, demonstrating its effectiveness in enhancing interpretability for medical image classification tasks.

This paper introduces a new, model-independent, metric, called RExQUAL, for quantifying the quality of explanations provided by attribution-based explainable artificial intelligence techniques and compare them. The underlying idea is based on feature attribution, using a subset of the ranking of the attributes highlighted by a model-agnostic explainable method in a forecasting task. Then, association rules are generated using these key attributes as input data. Novel metrics, including global support and confidence, are proposed to assess the joint quality of generated rules. Finally, the quality of the explanations is calculated based on a wise and comprehensive combination of the association rules global metrics. The proposed method integrates local explanations through attribution-based approaches for evaluation and feature selection with global explanations for the entire dataset. This paper rigorously evaluates the new metric by comparing three explainability techniques: the widely used SHAP and LIME, and the novel methodology RULEx. The experimental design includes predicting time series of different natures, including univariate and multivariate, through deep learning models. The results underscore the efficacy and versatility of the proposed methodology as a quantitative framework for evaluating and comparing explainable techniques.

Several interrelated variables typically characterize real-world processes, and a time series cannot be predicted without considering the influence that other time series might have on the target time series. This work proposes a novel algorithm to forecast multivariate big data time series. This new general-purpose approach consists first of a previous pattern recognition performed jointly using all time series that form the multivariate time series and then predicts the target time series by searching for similarities between pattern sequences. The proposed algorithm is designed to tackle multivariate time series forecasting problems within the context of big data. In particular, the algorithm has been developed with a distributed nature to enhance its efficiency in analyzing and processing large volumes of data. Moreover, the algorithm is straightforward to use, with only two parameters needing adjustment. Another advantage of the MV-bigPSF algorithm is its ability to perform multi-step forecasting, which is particularly useful in many practical applications. To evaluate the algorithm’s performance, real-world data from Uruguay’s power consumption has been utilized. Specifically, MV-bigPSF has been compared with both univariate and multivariate methods. Regarding the univariate ones, MV-bigPSF improved 12.8% in MAPE compared to the second-best method. Regarding the multivariate comparison, MV-bigPSF improved 44.8% in MAPE with respect to the second most accurate method. Regarding efficiency, the execution time of MV-bigPSF was 1.83 times faster than the second-fastest multivariate method, both in a single-core environment. Therefore, the proposed algorithm can be a valuable tool for practitioners and researchers working in multivariate time series forecasting, particularly in big data applications.

Time series is one of the most common data types in the industry nowadays. Forecasting the future of a time series behavior can be useful in planning ahead, saving time, resources, and helping avoid undesired scenarios. To make the forecasting, historical data is utilized due to the causal nature of the time series. Several deep learning algorithms have been presented in this area, where the input is processed through a series of non-linear functions to produce the output. We present a novel strategy to improve the performance of deep learning models in time series forecasting in terms of efficiency while reaching similar effectiveness. This approach separates the model into levels, starting with the easiest and continuing to the most difficult. The simpler levels deal with smoothed versions of the input, whereas the most sophisticated level deals with the raw data. This strategy seeks to mimic the human learning process, in which basic tasks are completed initially, followed by more precise and sophisticated ones. Our method achieved promising results, obtaining a 35% improvement in mean squared error and a 2.6 time decrease in training time compared with the best models found in a variety of time series.

Machine and deep learning has become one of the most useful tools in the last years as a diagnosis-decision-support tool in the health area. However, it is widely known that artificial intelligence models are considered a black box and most experts experience difficulties explaining and interpreting the models and their results. In this context, explainable artificial intelligence is emerging with the aim of providing black-box models with sufficient interpretability so that models can be easily understood and further applied. Obstructive sleep apnea is a common chronic respiratory disease related to sleep. Its diagnosis nowadays is done by processing different data signals, such as electrocardiogram or respiratory rate. The waveform of the respiratory signal is of importance too. Machine learning models could be applied to the signal's analysis. Data from a polysomnography study for automatic sleep apnea detection have been used to evaluate the use of the Local Interpretable Model-Agnostic (LIME) library for explaining the health data models. Results obtained help to understand how several features have been used in the model and their influence in the quality of sleep.

Precision agriculture focuses on the development of site-specific harvest considering the variability of each crop area. Vegetation indices allow the study and delineation of different characteristics of each field zone, generally invisible to the naked-eye. This paper introduces a new big data triclustering approach based on evolutionary algorithms. The algorithm shows its capability to discover three-dimensional patterns on the basis of vegetation indices from vine crops. Different vegetation indices have been tested to find different patterns in the crops. The results reported using a vineyard crop located in Portugal depicts four areas with different moisture stress particularities that can lead to changes in the management of the vineyard. Furthermore, scalability studies have been performed, showing that the proposed algorithm is suitable for dealing with big datasets.

The agricultural sector has been, and still is, the most important economic sector in many countries. Due to advances in technology, the amount and variety of available data have been increasing over the years. However, compared to other economic sectors, there is not always enough quality data for one particular domain (crops, plantations, plots) to obtain acceptable forecasting results with machine learning algorithms. In this context, transfer learning can help extract knowledge from different but related domains with enough data to transfer it to a target domain with scarce data. This process can overcome forecasting accuracy compared to training models uniquely with data from the target domain. In this work, a novel instance weighting-based transfer learning algorithm is proposed and applied to the phenology forecasting problem. A new metric named DIAFAN is proposed to weight samples from different source domains according to their relationship with the target domain, promoting the diversity of the information and avoiding inconsistent samples. Additionally, a set of validation schemes is specifically designed to ensure fair comparisons in terms of data volume with other benchmark transfer learning algorithms. The proposed algorithm, DIAFAN-TL, is tested with a proposed dataset of 16 plots of olive groves from different places, including information fusion from satellite images, meteorological stations and human field sampling of crop phenology. DIAFAN-TL achieves a remarkable improvement with respect to 15 other well-known transfer learning algorithms and three nontransfer learning scenarios. Finally, several performance analyses according to the different phenological states, prediction horizons and source domains are also performed.

Nowadays, electricity is a basic commodity necessary for the well-being of any modern society. Due to the growth in electricity consumption in recent years, mainly in large cities, electricity forecasting is key to the management of an efficient, sustainable and safe smart grid for the consumer. In this work, a deep neural network is proposed to address the electricity consumption forecasting in the short-term, namely, a long short-term memory (LSTM) network due to its ability to deal with sequential data such as time-series data. First, the optimal values for certain hyper-parameters have been obtained by a random search and a metaheuristic, called coronavirus optimization algorithm (CVOA), based on the propagation of the SARS-Cov-2 virus. Then, the optimal LSTM has been applied to predict the electricity demand with 4-h forecast horizon. Results using Spanish electricity data during nine years and half measured with 10-min frequency are presented and discussed. Finally, the performance of the proposed LSTM using random search and the LSTM using CVOA is compared, on the one hand, with that of recently published deep neural networks (such as a deep feed-forward neural network optimized with a grid search) and temporal fusion transformers optimized with a sampling algorithm, and, on the other hand, with traditional machine learning techniques, such as a linear regression, decision trees and tree-based ensemble techniques (gradient-boosted trees and random forest), achieving the smallest prediction error below 1.5%.

Abstract In this paper, we consider the task of predicting the electricity power generated by photovoltaic solar systems for the next day at half‐hourly intervals. We introduce DL, a deep learning approach based on feed‐forward neural networks for big data time series, which decomposes the forecasting problem into several sub‐problems. We conduct a comprehensive evaluation using 2 years of Australian solar data, evaluating accuracy and training time, and comparing the performance of DL with two other advanced methods based on neural networks and pattern sequence similarity. We investigate the use of multiple data sources (solar power and weather data for the previous days, and weather forecast for the next day) and also study the effect of different historical window sizes. The results show that DL produces competitive accuracy results and scales well, and is thus a highly suitable method for big data environments.

Accurate forecasting of the power generated by PhotoVoltaic (PV) systems is needed for the successful integration of solar power into the electricity grid. In this paper, we consider the task of simultaneously forecasting the PV power output for the next day at half-hourly intervals using only previous PV power data. We propose a number of methods for constructing static and dynamic ensembles of neural networks. The static ensembles are based on random sampling and random feature selection, and the dynamic ones adaptively weight the contribution of the ensemble members based on their recent performance. We conduct an evaluation using Australian solar PV data for two years, and compare the results with state-of-the-art single prediction models and classical ensemble models. The results show that the proposed static ensembles are beneficial, achieving higher accuracy than the single neural networks and the other single and ensemble models used for comparison. The dynamic ensemble versions further improve the accuracy, demonstrating the potential of adaptively combining the predictions of ensemble members for accurate solar power forecasting.

We present PSF-NN, a new approach for time series forecasting. It combines prediction based on sequence similarity with neural networks. PSF-NN first generates predictions using the PSF algorithm that are then refined by the neural network component, which also utilizes additional features. We evaluate the performance of PSF-NN using a time series of hourly electricity demands for the state of New South Wales in Australia for three years. The task is to predict an interval of future values simultaneously, i.e. the 24 demands for the next day, instead of predicting just a single future demand. The results showed that the combined PSF-NN approach provides accurate predictions, outperforming the original PSF algorithm and a number of baselines.

Many approaches are currently devoted to find DNA motifs in nucleotide sequences. However, this task remains challenging for specialists nowadays due to the difficulties they find to deeply understand gene regulatory mechanisms, especially when analyzing binding sites in DNA. These sites or specific nucleotide sequences are known to be responsible for transcription processes. Thus, this work aims at providing an updated overview on strategies developed to discover meaningful motifs in DNA-related sequences, and, in particular, their attempts to find out relevant binding sites. From all existing approaches, this work is focused on dictionary, ensemble, and artificial intelligence-based algorithms since they represent the classical and the leading ones, respectively.

This paper presents a new approach to forecast the behavior of time series based on similarity of pattern sequences. First, clustering techniques are used with the aim of grouping and labeling the samples from a data set. Thus, the prediction of a data point is provided as follows: first, the pattern sequence prior to the day to be predicted is extracted. Then, this sequence is searched in the historical data and the prediction is calculated by averaging all the samples immediately after the matched sequence. The main novelty is that only the labels associated with each pattern are considered to forecast the future behavior of the time series, avoiding the use of real values of the time series until the last step of the prediction process. Results from several energy time series are reported and the performance of the proposed method is compared to that of recently published techniques showing a remarkable improvement in the prediction.

Biomedical research has been revolutionized by high-throughput techniques and the enormous amount of bio-logical data they are able to generate. In particular micro-array technology has the capacity to monitor changes in RNA abundance for thousands of genes simultaneously. The interest shown over microarray analysis methods has rapidly raised. Clustering is widely used in the analysis of microarray data to group genes of interest targeted from microarray experiments on the basis of similarity of expression patterns. In this work we apply two clustering algorithms, K-means and Expectation Maximization to particular a problem and we compare the groupings obtained on the basis of the cohesiveness of the gene products associated to the genes in each cluster.

A new approach is presented in this work with the aim of predicting time series behaviors. A previous labeling of the samples is obtained utilizing clustering techniques and the forecasting is applied using the information provided by the clustering. Thus, the whole data set is discretized with the labels assigned to each data point and the main novelty is that only these labels are used to predict the future behavior of the time series, avoiding using the real values of the time series until the process ends. The results returned by the algorithm, however, are not labels but the nominal value of the point that is required to be predicted. The algorithm based on labeled (LBF) has been tested in several energy-related time series and a notable improvement in the prediction has been achieved.

Breast cancer is one of the diseases causing the largest number of deaths among women. Its early detection has been proved to be the most effective way to combat it. This work is focused on developing an integral tool able to detect microcalcifications in mammographies, since the presence of these particles is a clear symptom of an incipient cancer. The proposed approach combines two techniques successfully used in other areas separately, such as linear pixel prediction and support-vector machines, in order to obtain almost perfect prediction accuracy. Moreover, a filter has been designed with the aim of decrease the processing time. The result verges on 96% of hits, improving previous works by 6%, on average.