Search Results (20)

The Slime Mould Algorithm (SMA) is a widely used swarm intelligence algorithm. Encouraged by the theory of no free lunch and the inherent shortcomings of the SMA, this work proposes a new variant of the SMA, called the BWSMA, in which three improvement mechanisms are integrated. The adaptive greedy mechanism is used to accelerate the convergence of the algorithm and avoid ineffective updates. The best–worst management strategy improves the quality of the population and increases its search capability. The stagnant replacement mechanism prevents the algorithm from falling into a local optimum by replacing stalled individuals. In order to verify the effectiveness of the proposed method, this paper conducts a full range of experiments on the CEC2018 test suite and the CEC2022 test suite and compares BWSMA with three derived algorithms, eight SMA variants, and eight other improved algorithms. The experimental results are analyzed using the Wilcoxon rank-sum test, the Friedman test, and the Nemenyi test. The results indicate that the BWSMA significantly outperforms these compared algorithms. In the comparison with the SMA variants, the BWSMA obtained average rankings of 1.414, 1.138, 1.069, and 1.414. In comparison with other improved algorithms, the BWSMA obtained average rankings of 2.583 and 1.833. Finally, the applicability of the BWSMA is further validated through two structural optimization problems. In conclusion, the proposed BWSMA is a promising algorithm with excellent search accuracy and robustness. Full article

This study addresses the critical need for effective data fusion strategies in pest prediction for pitahaya (dragon fruit) cultivation in the Ecuadorian Amazon, where heterogeneous data sources—such as environmental sensors and chlorophyll measurements—offer complementary but fragmented insights. Current agricultural monitoring systems often fail to integrate these data streams, limiting early pest detection accuracy. To overcome this, we compared early and late fusion approaches using comprehensive experiments. Multidimensionality is a central challenge: the datasets span temporal (hourly sensor readings), spatial (plot-level chlorophyll samples), and spectral (chlorophyll reflectance) dimensions. We applied dimensionality reduction techniques—PCA, KPCA (linear, polynomial, RBF), t-SNE, and UMAP—to preserve relevant structure and enhance interpretability. Evaluation metrics included the proportion of information retained (score) and cluster separability (silhouette score). Our results demonstrate that early fusion yields superior integrated representations, with PCA and KPCA-linear achieving the highest scores (0.96 vs. 0.94), and KPCA-poly achieving the best cluster definition (silhouette: 0.32 vs. 0.31). Statistical validation using the Friedman test (${\chi }^2$ = 12.00, p = 0.02) and Nemenyi post hoc comparisons (p < 0.05) confirmed significant performance differences. KPCA-RBF performed poorly (score: 0.83; silhouette: 0.05), and although t-SNE and UMAP offered visual insights, they underperformed in clustering (silhouette < 0.12). These findings make three key contributions. First, early fusion better captures cross-domain interactions before dimensionality reduction, improving prediction robustness. Second, KPCA-poly offers an effective non-linear mapping suitable for tropical agroecosystem complexity. Third, our framework, when deployed in Joya de los Sachas, improved pest prediction accuracy by 12.60% over manual inspection, leading to more targeted pesticide use. This contributes to precision agriculture by providing low-cost, scalable strategies for smallholder farmers. Future work will explore hybrid fusion pipelines and sensor-agnostic models to extend generalizability. Full article

This study conducts an in-depth investigation of the performance of six transformer models using 12 different datasets—10 with three classes and two with two classes—on sentiment classification. We use these six models and generate all combinations of triple schema ensembles, Majority and Soft vote. In total, we compare 46 classifiers on each dataset and see in one case up to a 7.6% increase in accuracy on a dataset with three classes from an ensemble scheme and, in a second case, up to 8.5% increase in accuracy on a dataset with two classes. Our study contributes to the field of natural language processing by exploring the reasons for the predominance, in this particular task, of Majority vote over Soft vote. The conclusions are drawn after a thorough investigation of the classifiers that are co-compared with each other through reliability charts, analyses of the confidence the models have in their predictions and their metrics, concluding with statistical analyses using the Friedman test and the Nemenyi post-hoc test with useful conclusions. Full article

This study examines the efficacy of Random Forest and XGBoost classifiers in conjunction with three upsampling techniques—SMOTE, ADASYN, and Gaussian noise upsampling (GNUS)—across datasets with varying class imbalance levels, ranging from moderate to extreme (15% to 1% churn rate). Employing metrics such as F1 score, ROC AUC, PR AUC, Matthews Correlation Coefficient (MCC), and Cohen’s Kappa, this research provides a comprehensive evaluation of classifier performance under different imbalance scenarios, focusing on applications in the telecommunications domain. The findings highlight that tuned XGBoost paired with SMOTE (Tuned_XGB_SMOTE) consistently achieves the highest F1 score and robust performance across all imbalance levels. SMOTE emerged as the most effective upsampling method, particularly when used with XGBoost, whereas Random Forest performed poorly under severe imbalance. ADASYN showed moderate effectiveness with XGBoost but underperformed with Random Forest, and GNUS produced inconsistent results. This study underscores the impact of data imbalance, with MCC, Kappa, and F1 scores fluctuating significantly, whereas ROC AUC and PR AUC remained relatively stable. Moreover, rigorous statistical analyses employing the Friedman test and Nemenyi post hoc comparisons confirmed that the observed improvements in F1 score, PR-AUC, Kappa, and MCC were statistically significant (p < 0.05), with Tuned_XGB_SMOTE significantly outperforming Tuned_RF_GNUS. While differences in ROC-AUC were not significant, the consistency of these results across multiple performance metrics underscores the reliability of our framework, offering a statistically validated and attractive solution for model selection in imbalanced classification scenarios. Full article

Classifying imbalanced data is a well-known challenge in machine learning. One of the fields inherently affected by imbalanced data is credit datasets in finance. In this study, to address this challenge, we employed one of the most recent methods developed for classifying imbalanced data, CorrOV-CSEn. In addition to the original CorrOV-CSEn approach, which uses AdaBoost as its base learning method, we also applied Multi-Layer Perceptron (MLP), random forest, gradient boosted trees, XGBoost, and CatBoost. Our dataset, sourced from the Iran capital market from 2015 to 2022, utilizes the more general and accurate term business failure instead of default. Model performance was evaluated using sensitivity, precision, and F1 score, while their overall performance was compared using the Friedman–Nemenyi test. The results indicate the high effectiveness of all models in identifying failing businesses (sensitivity), with CatBoost achieving a sensitivity of 0.909 on the test data. However, all models exhibited relatively low precision. Full article

This study examines the performance of various machine learning (ML) models in predicting Interstitial Glucose (IG) levels using data from wrist-worn wearable sensors. The insights from these predictions can aid in understanding metabolic syndromes and disease states. A public dataset comprising information from the Empatica E4 smart watch, the Dexcom Continuous Glucose Monitor (CGM) measuring IG, and a food log was utilized. The raw data were processed into features, which were then used to train different ML models. This study evaluates the performance of decision tree (DT), support vector machine (SVM), Random Forest (RF), Linear Discriminant Analysis (LDA), K-Nearest Neighbors (KNN), Gaussian Naïve Bayes (GNB), lasso cross-validation (LassoCV), Ridge, Elastic Net, and XGBoost models. For classification, IG labels were categorized into high, standard, and low, and the performance of the ML models was assessed using accuracy (40–78%), precision (41–78%), recall (39–77%), F1-score (0.31–0.77), and receiver operating characteristic (ROC) curves. Regression models predicting IG values were evaluated based on R-squared values (−7.84–0.84), mean absolute error (5.54–60.84 mg/dL), root mean square error (9.04–68.07 mg/dL), and visual methods like residual and QQ plots. To assess whether the differences between models were statistically significant, the Friedman test was carried out and was interpreted using the Nemenyi post hoc test. Tree-based models, particularly RF and DT, demonstrated superior accuracy for classification tasks in comparison to other models. For regression, the RF model achieved the lowest RMSE of 9.04 mg/dL with an R-squared value of 0.84, while the GNB model performed the worst, with an RMSE of 68.07 mg/dL. A SHAP analysis identified time from midnight as the most significant predictor. Partial dependence plots revealed complex feature interactions in the RF model, contrasting with the simpler interactions captured by LDA. Full article

Landslides represent one of the main geological hazards, especially in Chile. The main purpose of this study is to evaluate the application of machine learning algorithms (SVM, RF, XGBoost and logistic regression) and compare the results for the modeling of landslide susceptibility in the province of Chañaral, III region, Chile. A total of 86 sites are identified using various sources, in addition to 86 non-landslide sites. This spatial data management and analysis are conducted using QGIS software. The sites are randomly divided, and then a cross-validation process is applied to calculate the accuracy of the models. After that, from 22 conditioning factors, 12 are chosen based on the information gain ratio (IGR). Subsequently, five factors are excluded by the correlation criterion. After this analysis, two indices not previously utilized in the literature, the NDGI (normalized difference glacier index) and EVI (enhanced vegetation index), are employed for the final model. The performance of the models is evaluated through the area under the ROC (receiver operating characteristic) curve (AUC). To study the statistical behavior of the model, the Friedman nonparametric test is performed to compare the performance with the other algorithms and the Nemenyi test for pairwise comparison. Of the algorithms used, RF (AUC = 0.957) and XGBoost (AUC = 0.955) have the highest accuracy values measured in AUC compared to the other models and can be used for the same purpose in other geographic areas with similar characteristics. The findings of this investigation have the potential to assist in land use planning, landslide risk reduction, and informed decision making in the surrounding zones. Full article

Slope displacement is a crucial factor that affects slope stability in engineering construction. The monitoring and prediction of slope displacement are especially important to ensure slope stability. To achieve this goal, it is necessary to establish an effective prediction model and analyze the patterns and trends of slope displacement. In recent years, monitoring efforts for high slopes have increased. With the growing availability of means and data for slope monitoring, the accurate prediction of slope displacement accidents has become even more critical. However, the lack of an accurate and efficient algorithm has resulted in an underutilization of available data. In this paper, we propose a combined EEMD-IESSA-LSSVM algorithm. Firstly, we use EEMD to decompose the slope displacement data and then introduce a more efficient and improved version of the sparrow search algorithm, called the irrational escape sparrow search algorithm (IESSA), by optimizing it and incorporating adaptive weight factors. We compare the IESSA algorithm with SSA, CSSOA, PSO, and GWO algorithms through validation using three different sets of benchmark functions. This comparison demonstrates that the IESSA algorithm achieves higher accuracy and a faster solving speed in solving these functions. Finally, we optimize LSSVM to predict slope displacement by incorporating rainfall and water level data. To verify the reliability of the algorithm, we conduct simulation analysis using slope data from the xtGTX1 monitoring point and the xtGTX3 monitoring point in the Yangtze River Xin Tan landslide and compare the results with those obtained using EEMD-LSSVM, EEMD-SSA-LSSVM, and EEMD-GWO-LSSVM. After numerical simulation, the goodness-of-fit of the two monitoring points is 0.98998 and 0.97714, respectively, which is 42% and 34% better than before. Using Friedman and Nemenyi tests, the algorithms were ranked as follows: IESSA-LSSVM > GWO-LSSVM > SSA-LSSVM > LSSVM. The findings indicate that the combined EEMD-IESSA-LSSVM algorithm exhibits a superior prediction ability and provides more accurate predictions for slope landslides compared to other algorithms. Full article

Unmanned Aerial Vehicles (UAVs) are increasingly utilized for urban patrol and defense owing to their low cost, high mobility, and rapid deployment. This paper proposes a multi-UAV mission planning model that takes into account mission execution rates, flight energy consumption costs, and impact costs. A kinematics and dynamics model of a quadcopter UAV is established, and the UAV’s flight state is analyzed. Due to the difficulties in addressing 3D UAV kinematic constraints and poor uniformity using traditional optimization algorithms, a lightning search algorithm (LSA) based on multi-layer nesting and random walk strategies (MNRW-LSA) is proposed. The convergence performance of the MNRW-LSA algorithm is demonstrated by comparing it with several other algorithms, such as the Golden Jackal Optimization (GJO), Hunter–Prey Optimization (HPO), Pelican Optimization Algorithm (POA), Reptile Search Algorithm (RSA), and the Golden Eagle Optimization (GEO) using optimization test functions, Friedman and Nemenyi tests. Additionally, a greedy strategy is added to the Rapidly-Exploring Random Tree (RRT) algorithm to initialize the trajectories for simulation experiments using a 3D city model. The results indicate that the proposed algorithm can enhance global convergence and robustness, shorten convergence time, improve UAV execution coverage, and reduce energy consumption. Compared with other algorithms, such as Particle Swarm Optimization (PSO), Simulated Annealing (SA), and LSA, the proposed method has greater advantages in addressing multi-UAV trajectory planning problems. Full article

Intuitionistic fuzzy parameterized intuitionistic fuzzy soft matrices (ifpifs-matrices), proposed by Enginoğlu and Arslan in 2020, are worth utilizing in data classification in supervised learning due to coming into prominence with their ability to model decision-making problems. This study aims to define the concepts metrics, quasi-, semi-, and pseudo-metrics and similarities, quasi-, semi-, and pseudo-similarities over ifpifs-matrices; develop a new classifier by using them; and apply it to data classification. To this end, it develops a new classifier, i.e., Intuitionistic Fuzzy Parameterized Intuitionistic Fuzzy Soft Classifier (IFPIFSC), based on six pseudo-similarities proposed herein. Moreover, this study performs IFPIFSC’s simulations using 20 datasets provided in the UCI Machine Learning Repository and obtains its performance results via five performance metrics, accuracy (Acc), precision (Pre), recall (Rec), macro F-score (MacF), and micro F-score (MicF). It also compares the aforementioned results with those of 10 well-known fuzzy-based classifiers and 5 non-fuzzy-based classifiers. As a result, the mean Acc, Pre, Rec, MacF, and MicF results of IFPIFSC, in comparison with fuzzy-based classifiers, are 94.45%, 88.21%, 86.11%, 87.98%, and 89.62%, the best scores, respectively, and with non-fuzzy-based classifiers, are 94.34%, 88.02%, 85.86%, 87.65%, and 89.44%, the best scores, respectively. Later, this study conducts the statistical evaluations of the performance results using a non-parametric test (Friedman) and a post hoc test (Nemenyi). The critical diagrams of the Nemenyi test manifest the performance differences between the average rankings of IFPIFSC and 10 of the 15 are greater than the critical distance (4.0798). Consequently, IFPIFSC is a convenient method for data classification. Finally, to present opportunities for further research, this study discusses the applications of ifpifs-matrices for machine learning and how to improve IFPIFSC. Full article

Improving the accuracy and stability of daily runoff prediction is crucial for effective water resource management and flood control. This study proposed a novel stacking ensemble learning model based on attention mechanism for the daily runoff prediction. The proposed model has a two-layer structure with the base model and the meta model. Three machine learning models, namely random forest (RF), adaptive boosting (AdaBoost), and extreme gradient boosting (XGB) are used as the base models. The attention mechanism is used as the meta model to integrate the output of the base model to obtain predictions. The proposed model is applied to predict the daily inflow to Fuchun River Reservoir in the Qiantang River basin. The results show that the proposed model outperforms the base models and other ensemble models in terms of prediction accuracy. Compared with the XGB and weighted averaging ensemble (WAE) models, the proposed model has a 10.22% and 8.54% increase in Nash–Sutcliffe efficiency (NSE), an 18.52% and 16.38% reduction in root mean square error (RMSE), a 28.17% and 18.66% reduction in mean absolute error (MAE), and a 4.54% and 4.19% increase in correlation coefficient (r). The proposed model significantly outperforms the base model and simple stacking model indicated by both the Friedman test and the Nemenyi test. Thus, the proposed model can produce reasonable and accurate prediction of the reservoir inflow, which is of great strategic significance and application value in formulating the rational allocation and optimal operation of water resources and improving the breadth and depth of hydrological forecasting integrated services. Full article

The electroencephalogram (EEG) can objectively reflect the emotional state of human beings, and has attracted much attention in the academic circles in recent years. However, due to its weak, non-stationary, and low signal-to-noise properties, it is inclined to cause noise in the collected EEG data. In addition, EEG features extracted from different frequency bands and channels usually exhibit different levels of emotional expression abilities in emotion recognition tasks. In this paper, we fully consider the characteristics of EEG and propose a new model RSRRW (retargeted semi-supervised regression with robust weights). The advantages of the new model can be listed as follows. (1) The probability weight is added to each sample so that it could help effectively search noisy samples in the dataset, and lower the effect of them at the same time. (2) The distance between samples from different categories is much wider than before by extending the $ϵ$-dragging method to a semi-supervised paradigm. (3) Automatically discover the EEG emotional activation mode by adaptively measuring the contribution of sample features through feature weights. In the three cross-session emotion recognition tasks, the average accuracy of the RSRRW model is 81.51%, which can be seen in the experimental results on the SEED-IV dataset. In addition, with the support of the Friedman test and Nemenyi test, the classification of RSRRW model is much more accurate than that of other models. Full article

Machine learning models may not be able to effectively learn and predict from imbalanced data in the fields of machine learning and data mining. This study proposed a method for analyzing the performance impact of imbalanced binary data on machine learning models. It systematically analyzes 1. the relationship between varying performance in machine learning models and imbalance rate (IR); 2. the performance stability of machine learning models on imbalanced binary data. In the proposed method, the imbalanced data augmentation algorithms are first designed to obtain the imbalanced dataset with gradually varying IR. Then, in order to obtain more objective classification results, the evaluation metric AFG, arithmetic mean of area under the receiver operating characteristic curve (AUC), F-measure and G-mean are used to evaluate the classification performance of machine learning models. Finally, based on AFG and coefficient of variation (CV), the performance stability evaluation method of machine learning models is proposed. Experiments of eight widely used machine learning models on 48 different imbalanced datasets demonstrate that the classification performance of machine learning models decreases with the increase of IR on the same imbalanced data. Meanwhile, the classification performances of LR, DT and SVC are unstable, while GNB, BNB, KNN, RF and GBDT are relatively stable and not susceptible to imbalanced data. In particular, the BNB has the most stable classification performance. The Friedman and Nemenyi post hoc statistical tests also confirmed this result. The SMOTE method is used in oversampling-based imbalanced data augmentation, and determining whether other oversampling methods can obtain consistent results needs further research. In the future, an imbalanced data augmentation algorithm based on undersampling and hybrid sampling should be used to analyze the performance impact of imbalanced binary data on machine learning models. Full article

In this work, the performance of five deep learning architectures in classifying COVID-19 in a multi-class set-up is evaluated. The classifiers were built on pretrained ResNet-50, ResNet-50r (with kernel size $5\times 5$ in the first convolutional layer), DenseNet-121, MobileNet-v3 and the state-of-the-art CaiT-24-XXS-224 (CaiT) transformer. The cross entropy and weighted cross entropy were minimised with Adam and AdamW. In total, 20 experiments were conducted with 10 repetitions and obtained the following metrics: accuracy (Acc), balanced accuracy (BA), F₁ and F₂ from the general F${}_{\beta }$ macro score, Matthew’s Correlation Coefficient (MCC), sensitivity (Sens) and specificity (Spec) followed by bootstrapping. The performance of the classifiers was compared by using the Friedman–Nemenyi test. The results show that less complex architectures such as ResNet-50, ResNet-50r and DenseNet-121 were able to achieve better generalization with rankings of 1.53, 1.71 and 3.05 for the Matthew Correlation Coefficient, respectively, while MobileNet-v3 and CaiT obtained rankings of 3.72 and 5.0, respectively. Full article

Recent advances in stroke treatment have provided effective tools to successfully treat ischemic stroke, but still a majority of patients are not treated due to late arrival to hospital. With modern stroke treatment, earlier arrival would greatly improve the overall treatment results. This prospective study was performed to asses the capability of bilateral accelerometers worn in bracelets 24/7 to detect unilateral arm paralysis, a hallmark symptom of stroke, early enough to receive treatment. Classical machine learning algorithms as well as state-of-the-art deep neural networks were evaluated on detection times between 15 min and 120 min. Motion data were collected using triaxial accelerometer bracelets worn on both arms for 24 h. Eighty-four stroke patients with unilateral arm motor impairment and 101 healthy subjects participated in the study. Accelerometer data were divided into data windows of different lengths and analyzed using multiple machine learning algorithms. The results show that all algorithms performed well in separating the two groups early enough to be clinically relevant, based on wrist-worn accelerometers. The two evaluated deep learning models, fully convolutional network and InceptionTime, performed better than the classical machine learning models with an AUC score between 0.947–0.957 on 15 min data windows and up to 0.993–0.994 on 120 min data windows. Window lengths longer than 90 min only marginally improved performance. The difference in performance between the deep learning models and the classical models was statistically significant according to a non-parametric Friedman test followed by a post-hoc Nemenyi test. Introduction of wearable stroke detection devices may dramatically increase the portion of stroke patients eligible for revascularization and shorten the time to treatment. Since the treatment effect is highly time-dependent, early stroke detection may dramatically improve stroke outcomes. Full article