Accurate prediction of the Remaining Useful Life of lithium-ion batteries is essential for enhancing reliability, safety, and maintenance planning in energy storage systems. This study proposes an optimized hybrid deep learning framework that integrates Convolutional Neural Networks with Deep Neural Networks for effective battery remaining useful life prediction. The model leverages convolutional neural network layers to automatically extract discriminative features from multivariate battery degradation data, while deep neural network layers model complex nonlinear relationships for precise regression estimation. To further enhance predictive performance and reduce feature redundancy, Binary Particle Swarm Optimization is employed for optimal feature selection. Experiments were conducted using a publicly available lithium-ion battery degradation dataset comprising 680 samples with electrical, thermal, and health-related parameters, including State of Health and remaining useful life indicators. The dataset was divided into training, validation, and testing subsets with proportions of 70%, 15%, and 15%, respectively. A comprehensive performance evaluation was performed using Mean Squared Error, Mean Absolute Percentage Error, Median Absolute Error, Mean Absolute Error, and the coefficient of determination. The proposed hybrid model achieved superior performance with a Mean Squared Error of 0.0141, Mean Absolute Error of 0.0931, Mean Absolute Percentage Error of 0.0142, Median Absolute Error of 0.0739, and a coefficient of determination of 99.01%, significantly outperforming comparative deep learning models. These results demonstrate that the proposed framework provides a robust and accurate solution for lithium-ion battery remaining useful life prediction and supports its potential deployment in intelligent battery management systems.

An unsupervised hybrid feature selection approach using scalable laplacian score and binary particle swarm optimization

Type 1 Diabetes Mellitus (T1D) has been increasingly associated with alterations in the gut microbiome. However, the impact of taxonomic resolution, feature selection strategies, and machine learning methods on microbiome-based prediction remains incompletely understood. We analyzed publicly available 16S rRNA gene sequencing datasets from two geographic cohorts to evaluate microbiome-based prediction of T1D. Microbial features were constructed at multiple taxonomic levels and as full hierarchical taxonomic paths preserving phylogenetic structure. Machine learning models were trained using stratified cross-validation and cross-cohort validation frameworks. Feature selection was performed using Binary Particle Swarm Optimization (BPSO) to identify compact and predictive microbial signatures. Model performance was evaluated using AUC, Accuracy, F1 score, and Matthews Correlation Coefficient. Differential abundance analysis using the LinDA framework was used to support biological interpretation of selected taxa. Tree-based models, particularly Random Forest and XGBoost, achieved the strongest predictive performance across taxonomic representations. Taxonomic resolution influenced model behavior, with family-level features providing strong performance with compact feature sets, while higher-resolution representations did not consistently improve performance despite increased complexity. BPSO identified consistently selected taxa across validation frameworks, suggesting stable predictive signatures. Several of these taxa have been linked to inflammatory or metabolically altered gut environments. Cross-cohort validation showed reduced performance compared with within-study models, highlighting challenges in generalization. Machine learning combined with BPSO-based feature selection provides an effective framework for identifying predictive microbial signatures associated with T1D. Our findings highlight the importance of taxonomic resolution, feature stability, and cross-cohort validation in microbiome-based predictive modeling. Integrating evolutionary feature selection with machine learning and biological validation may improve the robustness and interpretability of candidate microbial signatures.

Software-Defined Networking (SDN) has emerged as a revolutionary paradigm. The integration of SDN within fog networks represents a synergistic convergence of two cutting-edge technologies. With the complexity of SDN serving fog networks, the optimization of communication cost becomes paramount. Addressing the intricate challenges of communication cost optimization necessitates the application of sophisticated methodologies. Multi-Objective Optimization (MOO) algorithms present a robust solution, allowing for the simultaneous optimization of multiple conflicting objectives. By employing MOO, this research proposes a bi-objective optimization model for the intra- and inter-domain communication cost of controller deployment in an SDN-based computing network. The evaluation performed has captured two aspects of the performance of using Binary Angle quantization Multi-objective Particle swarm optimization (BAMP) and Binary crowding Distance Angle quantization Multi-objective Particle swarm optimization (BDAMP) for SDN controllers’ deployment. The first aspect is multi-objective-based evaluation, and the second aspect is the SDN network performance. Our developed BAMP and BDAMP have shown superiority over the benchmarks in terms of both aspects. Most importantly, the best performance is achieved by BDAMP in terms of both intra- and inter- communication cost.

A Novel Personal Best- Control Binary Particle Swarm Optimization (NPbest-BPSO) based electromyography (EMG) signal feature selection and classification

Parkinson's disease (PD) affects 10million globally, with accurate staging essential for personalized treatment planning. Current UPDRS assessments achieve < 93% accuracy due to subjective clinical judgment and unimodal data limitations, failing to capture complex genetic-neuroimaging-clinical interactions driving disease heterogeneity. This study introduces MAFNet, a novel deep learning framework pioneering Iterative Adaptive Vold-Kalman Filter (IAVKF) temporal denoising, Accelerated Binary Particle Swarm Optimization (ABPSO) swarm feature selection, Multilayer Perceptron-Lagrangian Support Vector Machine (MLP-LSVM) classification, and Graph-Attention Based Multimodal Fusion Network (GAMF). Applied to PPMI cohort (200 patients) with genetic SNPs (50), neuroimaging voxels (1,024), and UPDRS-III scores, the end-to-end pipeline delivers 97.6% accuracy, 98.2% precision, 96.8% recall, and 97.3% F1-score-outperforming CNN (92.4%), Autoencoder (90.8%), InceptoFormer (96.6%), and HCT (97.0%). IAVKF boosts SNR + 15.2dB (+ 2.9% accuracy vs. PCA/t-SNE); ABPSO reduces 1,276→340 features (73% reduction); regularization cuts overfitting gap to 0.9% (vs. 4.2% baseline). SHAP interpretability validates clinical plausibility (top predictors: LRRK2 SNPs, UPDRS-III tremor, hippocampal volume). Five-fold CV confirms stability with the Indian cohort external validation. Real-time inference (0.2s/patient, RTX 3090) enables clinical deployment. Future scope includes longitudinal temporal modelling, modality-agnostic fusion, edge deployment, federated learning, and extension to Alzheimer's/ALS. MAFNet transforms PD staging from subjective assessments to objective precision medicine, enabling biomarker discovery, progression forecasting, and personalized therapies across diverse global populations.

ABSTRACT This paper presents an Adaptive Elite Hybrid Binary Particle Swarm–Grey Wolf Optimizer (AE-BPSO-BGWO) for the simultaneous optimization of distribution system reconfiguration (DSR), distributed generator (DG) sizing and placement, and electric vehicle (EV) charger allocation. The algorithm combines the exploration capability of Binary Particle Swarm Optimization with the exploitation strength of Binary Grey Wolf Optimizer and introduces an elite-adaptive mechanism that dynamically adjusts the search process in binary solution spaces. The performance of the proposed method is evaluated in two stages. First, benchmark tests are conducted on Sphere, Ackley, Griewank, Rosenbrock, Rastrigin, Schwefel, Zakharov, Levy, Michalewicz, and Bent Cigar functions, showing superior convergence speed and robustness compared with conventional and hybrid metaheuristic algorithms. Second, the algorithm is applied to IEEE 33- and 69-bus distribution systems under six scenarios: base case, reconfiguration only, DG allocation only, DG allocation after reconfiguration, reconfiguration after DG allocation, and simultaneous reconfiguration with DG allocation. In the 33-bus system, the reconfiguration-after-DG scenario achieves the lowest power loss of 38.28 kW and a minimum voltage of 0.9861 p.u. In the 69-bus system, the simultaneous optimization scenario reduces power loss to 27.16 kW with a minimum voltage of 0.9796 p.u., confirming the effectiveness of the proposed method.

Optimization of NIRS-based models for predicting quality and gas production traits of fresh alfalfa silage via machine learning.

Accurate and non-destructive assessment of fruit maturity is critical for sustainable agricultural practices. This study proposes a novel framework for evaluating strawberry ripeness using Mid-Infrared (MIR) spectroscopy combined with metaheuristic feature selection and supervised classification. A dataset of 443 strawberries spanning eight maturity stages was analyzed using six metaheuristic algorithms—Binary Grey Wolf Optimizer, Binary Particle Swarm Optimizer, Bee Colony Optimizer, Genetic Algorithm, Ant Colony Optimizer, and Gravitational Search Optimizer—integrated with four classifiers: Naïve Bayes, Decision Tree, Linear Discriminant Analysis, and Support Vector Machine. A new fitness function was designed to optimize classifier performance, and results were validated through Self-Organizing Map Neural Networks, cross-validation, and statistical significance testing. The Genetic Algorithm–Linear Discriminant Analysis combination achieved the highest and most stable accuracy (94.6–99%), outperforming existing image-based, deep learning, and conventional spectroscopic approaches while retaining interpretability. These findings demonstrate that metaheuristic-driven MIR analysis provides a robust, explainable, and efficient method for precise strawberry maturity assessment, offering significant potential for advancing eco-friendly and intelligent agricultural practices.

Early identification of lung cancer using questionnaire-based data offers a low-cost, non-invasive pathway to assist clinical decision-making. However, such datasets often contain redundant, noisy, and imbalanced attributes that limit the performance of traditional classifiers. This study introduces a hybrid LSTM-GRU framework optimized using a Grey Wolf-Whale Optimization (GWO-WOA) algorithm for hyperparameter tuning and Binary Particle Swarm Optimization (BPSO) for feature selection. Two public lung cancer datasets sourced from the Kaggle repository were employed: the first comprising 309 samples and the second containing 3000 samples. For both datasets, the preprocessing pipeline included missing-value imputation, categorical encoding, outlier removal, and z-score normalization to ensure feature consistency. Datasets were then split into 70%, 20%, and 10% subsets for training, validation, and testing, respectively. BPSO effectively selected the most informative features that contribute to accurate diagnosis. At the same time, GWO-WOA refined key hyperparameters, such as the learning rate, hidden units, and layer depth, of the hybrid architecture. Experimental results demonstrate the superior performance of the proposed GWO-WOA-LSTM-GRU model, achieving 100.00% accuracy, precision, recall, and F1-score on the 309-sample dataset, and 99.33% accuracy/F1 (precision: 99.34%, recall: 99.33%) on the 3000-sample dataset. In comparison, tuned single models-LSTM, GRU, CNN, and SVM-achieved accuracies ranging from 77.42 to 98.33%. These findings confirm that integrating metaheuristic optimization and hybrid recurrent networks enhances the robustness and generalization capabilities of lung cancer classification systems across diverse datasets, offering a reliable tool for early detection and clinical risk stratification.

The development of internet-connected environments to facilitate people to handle a variety of tasks is being made possible by the Internet of Things (IoT). Technology advancements offer businesses a numerous conveniences and benefits. However, they provide hackers and intruders more opportunities to investigate and take advantage of different methods to get through the security of IoT networks. Therefore, the primary challenges with the IoT based environments is security. It is essential to secure computer and Internet of Things platforms against a variety of threats and attacks. Furthermore, traditional intrusion detection systems (IDS) frequently handle massive amounts of data from IoT networks that contain redundant and irrelevant data, which results in decreased accuracy and longer response times. Therefore, this paper proposed an IDS to detect different types of attacks on IoT networks by employing the CIC-IoT-2023 dataset. The most relevant IoT network features are extracted by integrating the Improved Binary Grey Wolf Optimisation (IBGWO) and Particle Swarm Optimisation (PSO). Additionally, the Random Forest (RF) algorithm is used to analyse the selected features, using several decision trees to produce an accurate response. Therefore, the suggested Hybrid IBGWO-PSO-RF model has significantly outperformed the other IDS algorithms with an accuracy of 99.96% for multi-class classification.

Single-phase-to-ground faults occur frequently in distribution networks, while traditional localization methods have limitations such as insufficient feature extraction and poor topological adaptability. To address these issues, this paper proposes a two-stage localization method that integrates the Node Classification Matrix (NCM) and an Improved Binary Particle Swarm Optimization (IBPSO) algorithm. The NCM achieves rapid initial localization, and the IBPSO performs error correction. This paper employs an IEEE 33-node standard distribution network model to design simulations covering scenarios with varying fault locations, multiple fault resistances, and different numbers of node distortions for validation. The results demonstrate that the proposed method achieves a fault location accuracy of 96%, which is 19% higher than that of the NCM alone and 2% higher than that of the IBPSO alone. Moreover, it maintains an accuracy of over 95% under scenarios of 1–3 node distortions, topological switching, and high-impedance faults, and is compatible with existing Feeder Terminal Unit (FTU) devices. This method effectively balances localization speed and robustness, providing a reliable solution for the rapid fault isolation of distribution network.

Alzheimer’s disease is a progressive neurodegenerative disorder for which early detection remains a significant challenge due to the complexity of clinical features and the high dimensionality of medical data. This study aims to improve the accuracy and reliability of Alzheimer’s disease detection by evaluating the performance of multiple machine learning algorithms integrated with intelligent feature selection strategies. Five classification models, Decision Tree, Naïve Bayes, Random Forest, Logistic Regression, and Deep Learning, were investigated under two experimental scenarios: without feature selection and with feature selection using Recursive Feature Elimination, Binary Particle Swarm Optimization, and Variance Threshold. Model performance was evaluated using K-fold cross-validation based on accuracy, precision, recall, and F1-score metrics. The results demonstrate that feature selection consistently enhances classification performance, particularly for conventional machine learning models such as Random Forest and Logistic Regression. Although the Deep Learning model achieves competitive accuracy, its reduced precision and F1-score indicate limitations when applied to reduced feature spaces. These findings highlight the importance of incorporating appropriate feature selection techniques to address data complexity and improve the effectiveness of early Alzheimer’s disease detection.

In this paper, we propose a full-duplex integrated sensing and communication (ISAC) system enabled by a movable antenna (MA). By leveraging the characteristic of MA that can increase the spatial diversity gain, the performance of the system can be enhanced. We formulate a problem of minimizing the total transmit power consumption via jointly optimizing the discrete position of MA elements, beamforming vectors, sensing signal covariance matrix and user transmit power. Given the significant coupling of optimization variables, the formulated problem presents a non-convex optimization challenge that poses difficulties for direct resolution. To address this challenging issue, the discrete binary particle swarm optimization (BPSO) algorithm framework is employed to solve the formulated problem. Specifically, the discrete positions of MA elements are first obtained by iteratively solving the fitness function. The difference-of-convex (DC) programming and successive convex approximation (SCA) are used to handle non-convex and rank-1 terms in the fitness function. Once the BPSO iteration is complete, the discrete positions of MA elements can be determined, and we can obtain the solutions for beamforming vectors, sensing signal covariance matrix and user transmit power. Numerical results demonstrate the superiority of the proposed system in reducing the total transmit power consumption compared with fixed antenna arrays.

A novel multi-objective antenna optimization method based on the surrogate model-assisted deep reinforcement learning (SADRL) is proposed. The method is divided into three stages: coarse topology optimization, surrogate model construction, and fine topology optimization. First, the adaptive variable fidelity electromagnetic (AVFEM) model is used to assist the improved binary particle swarm optimization (IBPSO) algorithm for coarse optimization of antenna topology. This stage provides an initial database for surrogate-model training and a high-quality initial solution for subsequent deep reinforcement learning (DRL) algorithm. Second, the Bayesian Convolutional Neural Networks (BCNN) is employed as an online surrogate model, aiming to provide a low-cost interactive environment for the DRL. Finally, the deep Q-network (DQN) is used to perform fine optimization of antenna topology. To validate the proposed method, a multi-objective optimization of a monopole antenna is conducted with objectives of omnidirectionality, operating bandwidth, and in‑band gain flatness. The optimized design provides an operating band that covers 3.3–3.8 GHz and 5.75–5.85 GHz, while maintaining realized gains of 1.89 ± 0.23 dBi and 1.35 ± 0.11 dBi across the target bands, the azimuthal gain ripple is less than 2.86 dBi. Compared with other optimization methods, the proposed SADRL achieves the target design with fewer electromagnetic (EM) simulations.

This study aims to enhance the forecasting performance of slope stability predictions by applying and comparing the Binary Particle Swarm Optimization (BPSO) coupled with Support Vector Machine (BPSO-SVM) models. The BPSO technique is utilized to select relevant features from the dataset, thereby improving the overall effectiveness of the predictive models. The research includes 108 slope stability examples, with the dataset split between 70% training and 30% validation. The dataset comprises seven input parameters: cohesiveness, slope angle, unit weight, angle of internal friction, slope height, pore water pressure coefficient, and factor of safety. The objective is to classify the slope status, turning the problem into a classification task. To obtain optimal hyper-parameters for the SVM model, Grid Search was exploited. The accuracy of the slope stability predictions given by several models was assessed using receiver operating characteristic (ROC) curves. The results indicate that the BPSO-SVM model outperforms the standalone SVM and BPSO models, serving as a robust computational tool capable of accurately predicting slope stability.

A binary particle swarm optimization with dual encoding mechanism for feature selection

Automatic text summarization (ATS) deals with compressing a long document into a shorter version, retaining the key ideas of the original document. It aims to tackle the problem of information overload resulting from the continuous creation of documents. ATS helps people make news feeds, meeting reports, summarized legal and financial documents, study tools, article outlines, social media analysis, and so on. Extractive text summarization (ETS) is a type of ATS that creates summaries (extracts) by choosing sentences from the source without modifying the semantics and structure of the sentences. The main problem in ETS is to select the appropriate sentences that will constitute a coherent, concise, and nonredundant summary that covers the entire document. The literature suggests many optimization-based algorithms for the selection problem. Some authors have used binary particle swarm optimization (BPSO), genetic algorithm (GA), and artificial bee colony (ABC) algorithm to address the selection problem. However, these algorithms suffer from the problem of premature or slow convergence. This article proposes an ETS algorithm using BPSO and masked GA (EBPGA). The novelty of the proposed system is the usage of the masked GA (MGA), which leverages the properties of the document, namely, coherence, redundancy, relevance, and coverage factors, along with domain-specific mutation and crossover to tackle the premature convergence of the BPSO. Thus, the EBPGA algorithm selects the best sentences covering all the topics of the original document to create the summary that is coherent, relevant, and nonredundant. The proposed algorithm was evaluated by producing summaries of size 20%, 30%, and 40%, of the original document, and compared with the existing algorithms. The proposed summarization algorithm obtained a Recall-Oriented Understudy for Gisting Evaluation (ROUGE) score that is comparable with the contemporary algorithms.

The rising incidence and mortality in bladder cancer (BC) underscore the importance of identifying asscociated features. Current reliance on haematuria as a primary indicator for BC proves inadequate. While mining electronic health records (EHRs) offer potential of identifying BC-related signals, traditional data-driven methods struggle with high-dimensional datasets. This study aims to uncover novel BC-associated clinical signals by developing Parsimony-driven cAtegory-balaNced binary Signal extractor for Primary Care EHRs (PanSPICE) tailored to extremely high-dimensional data linked from multi-centres. We collected BC cases and control patients (n=64,884) linked at patient-level from Welsh nationwide databases, yielding 48,261 features in primary care settings. The PanSPICE approach begins with information gain to pre-rank features, then applies Retentive Stickiness Binary Particle Swarm Optimisation (RSBPSO) combined with C5.0 classification tree to overcome computational barriers in feature selection. A two-layer optimisation treated clinical signals in care processes (POC), diagnoses (DIAG), and medications (MED) separately to prevent feature masking. A tailored fitness function for RSBPSO to simultaneously optimise model performance and feature sparsity. Associations of the selected features were interpreted using logistic regression models adjusted for deprivation indices. The PanSPICE identified 38 optimal features (AUC (area under the curve)=0.81, 95% CI: 0.80-0.82), including urinary tract infections (OR=2.19, 95% CI: 2.05-2.14) and inverse associations with stroke (OR=0.64, 95% CI: 0.54-0.74) and dementia (OR=0.25, 95% CI: 0.17-0.35). Gender stratification revealed female-specific urine glucose testing association (OR=1.24, 95% CI: 1.08-1.43). Certain medications, such as trimethoprim, were positively associated with BC, while others, including ramipril and prednisolone, showed protective effects. The PanSPICE enables efficient high-dimensional EHR analysis, revealing under-recognised potential BC risk profiles and protective comorbidities. Gender-specific differences in BC associations highlight the importance of gender-stratified analyses, while computational advances provide a template for EHR-based clinical discovery. Findings warrant further mechanistic research into neurological protective pathways.

Antenna selection is an appealing energy-efficient solution for multiple-input multiple-output (MIMO) systems. This paper handled the antenna selection problem in MIMO systems as a multi-class classification task, and propounded an antenna selection technique for energy efficiency (EE) maximization based on Fast-Adaptive Base Class-Boost (Fast-ABC-Boost) which boosts the classification performance of many “weak” classifiers, i.e. regression trees, to produce a powerful “committee” to make classification decision. Simulation results prove the superiority of Fast-ABC-Boost over the up-to-date learning method based on Deep Reinforcement Learning (DRL) and the conventional optimization driven Cyclic Binary Particle Swarm Optimization (CBPSO) approach in terms of EE performance with feasible complexity.