In this study an adjusting post-processing approach is implemented for improving intra-hourly forecasts of solar power and ramp events of PV solar power systems at different locations in the United States. This study also serves as an out-of-sample test to evaluate the performance of the adjusting approach with different locations and timescales. Thus, various individual intra-hourly forecasts of solar power are combined and adjusted by applying the adjusting approach. Both point and probabilistic forecasts of solar power are included. After that, solar power ramp event forecasting by the adjusting approach is carried out.

This work compared the performance of three methods for constructing a regional ensemble prediction system (EPS) for wind speed forecasts: dynamical downscaling, breeding of growth modes (BGM), and blending method. The Weather Research and Forecasting (WRF) model was used to downscale the European Centre for Medium-range Weather Forecast (ECMWF) EPS. In addition, as the BGM method needs observation data for generating scaling factors, an alternative method for generating scaling factors was proposed to eliminate dependence on observation data. One-month tests between October 1st and October 30th, 2020, were implemented to evaluate the performance of three methods in the Gansu province of China. The results demonstrate that the blending method outperforms the other two methods. Furthermore, the difference in performance is evident mainly in early forecast lead time and becomes negligible as forecast time increases.

This study shows the application of self-organizing maps (SOMs) to probabilistic forecasts of wind power generation and ramps in Japan. SOMs are applied to atmospheric variables obtained from atmospheric reanalysis over the region, thus deriving classified weather patterns (WPs). Probabilistic relationships are established between the synoptic-scale atmospheric variables over East Japan and the generation of regionally integrated wind power in East Japan. Medium-range probabilistic wind power predictions are derived by SOM, as analog ensembles based on the WPs of the multi-center ensemble forecasts. As this analog approach handles stochastic uncertainties effectively, probabilistic wind power forecasts are rapidly generated from a very large number of forecast ensembles. The use of a multi-model ensemble provides better results than a one-forecast model. The hybrid ensemble forecasts further improve the probabilistic predictability skill of wind power generation, as compared with non-hybrid methods. It is expected that long-term wind forecasts will provide better guidance to transmission grid operators. The advantage of this method is that it can include an interpretative analysis of meteorological factors for variations in renewable energy.

Taiwan is highly susceptible to global warming, experiencing a 1.4°C increase in air temperatures from 1911-2005, which is twice the average for the Northern Hemisphere. This has led to higher rates of respiratory and cardiovascular mortality. Accurately predicting maximum temperatures during the summer season is crucial, but numerical weather models become less accurate and more uncertain beyond five days. To improve forecast reliability, statistical post-processing is needed to address systematic errors. In September 2020, NOAA NCEP implemented the Global Ensemble Forecast System version 12 (GEFSv12) to help manage climate risks. This study developed a Hybrid statistical post-processing method that combines Artificial Neural Networks (ANN) and Quantile mapping (QQ) approaches to predict daily maximum temperatures and extremes in Taiwan during the summer season. The Hybrid technique, utilizing deep learning techniques, was applied to the GEFSv12 reforecast data and evaluated against ERA5 reanalysis. The Hybrid technique was the most effective among the three techniques tested. It had the lowest bias, RMSE, and highest correlation coefficient. It successfully reduced the warm bias and overestimation of Tmax extreme days. This led to improved prediction skills for all forecast lead times. Compared to ANN and QQ, the Hybrid method was more effective in predicting summer daily Tmax and its extremes on an extended-range time scale deterministic and ensemble probabilistic forecasts over Taiwan.

Accurate and timely rainfall prediction enhances productivity and can aid proper planning in sectors such as agriculture, health, transport and water resources. This study is aimed at improving rainfall prediction using ensemble methods. It first assesses the performance of six convective schemes (Kain–Fritsch (KF); Betts–Miller–Janji´c (BMJ); Grell–Fretas (GF); Grell 3D ensemble (G3); New–Tiedke (NT) and Grell–Devenyi (GD)) using the root mean square error (RMSE) and mean error (ME) focusing on the March–May 2013 rainfall period over Uganda. 18 ensemble members are generated from the three best performing convective schemes (i.e. KF, GF & G3). The performance of three ensemble methods (i.e. ensemble mean (EM); ensemble mean analogue (EMA) and multi–member analogue ensemble (MAEM)) is also analyzed using the RMSE and ME. The EM presented a smaller RMSE compared to individual schemes (EM:10.02; KF:23.96; BMJ:26.04; GF:25.85; G3:24.07; NT:29.13 & GD:26.27) and a better bias (EM:-1.28; KF:-1.62; BMJ:-4.04; GF:-3.90; G3:-3.62; NT:-5.41 & GD:-4.07). The EMA and MAEM presented 13 out of 21 stations & 17 out of 21 stations respectively with smaller RMSE compared to EM thus demonstrating additional improvement in predictive performance. The MAEM is a new approach proposed and described in the study.

In this paper, we introduce deboost, a Python library devoted to weighted distance ensembling of predictions for regression and classification tasks. Its backbone resides on the scikit-learn library for default models and data preprocessing functions. It offers flexible choices of models for the ensemble as long as they contain the predict method, like the models available from scikit-learn. deboost is released under the MIT open-source license and can be downloaded from the Python Package Index (PyPI) at https://pypi.org/project/deboost. The source scripts are also available on a GitHub repository at https://github.com/weihao94/DEBoost.

We propose a new ensemble classification algorithm, named Super Random Subspace Ensemble (Super RaSE), to tackle the sparse classification problem. The proposed algorithm is motivated by the Random Subspace Ensemble algorithm (RaSE). The RaSE method was shown to be a flexible framework that can be coupled with any existing base classification. However, the success of RaSE largely depends on the proper choice of the base classifier, which is unfortunately unknown to us. In this work, we show that Super RaSE avoids the need to choose a base classifier by randomly sampling a collection of classifiers together with the subspace. As a result, Super RaSE is more flexible and robust than RaSE. In addition to the vanilla Super RaSE, we also develop the iterative Super RaSE, which adaptively changes the base classifier distribution as well as the subspace distribution. We show the Super RaSE algorithm and its iterative version perform competitively for a wide range of simulated datasets and two real data examples. The new Super RaSE algorithm and its iterative version are implemented in a new version of the R package RaSEn.

Bootstrap resampling techinques, introduced by Efron and Rubin, can be presented in a general Bayesian framework, approximating the statistical distribution of a statistical functional φ(F), where F is a random distribution function. Efron’s and Rubin’s bootstrap procedures can be extended introducing an informative prior through the Proper Bayesian bootstrap. In this paper different bootstrap techniques are used and compared in predictive classification and regression models based on ensemble approaches, i.e. bagging models involving decision trees. Proper Bayesian bootstrap, proposed by Muliere and Secchi, is used to sample the posterior distribution over trees, introducing prior distributions on the covariates and the target variable. The results obtained are compared with respect to other competitive procedures employing different bootstrap techniques. The empirical analysis reports the results obtained on simulated and real data.

The conventional machine learning (ML) algorithms are continuously advancing and evolving at a fast-paced by introducing the novel learning algorithms. ML models are continually improving using hybridization and ensemble techniques to empower computation, functionality, robustness, and accuracy aspects of modeling. Currently, numerous hybrid and ensemble ML models have been introduced. However, they have not been surveyed in a comprehensive manner. This paper presents the state of the art of novel ML models and their performance and application domains through a novel taxonomy.

Energy demand forecasting is a fundamental aspect of modern energy management. It impacts resource planning, economic stability, environmental sustainability, and energy security. This importance is making it critical for countries worldwide, particularly in cases like Türkiye, where the energy dependency ratio is notably high. The goal of this study is to propose ensemble machine learning methods such as boosting, bagging, blending, and stacking with hyperparameter tuning and k-fold cross-validation, and investigate the application of these methods for predicting Türkiye's energy demand. This study utilizes population, GDP per capita, imports and exports as input parameters based on historical data from 1979 to 2021 in Türkiye. Eleven combinations of all predictor variables were analyzed, and the best one was selected. It was observed that a very high correlation exists among population, GDP, imports, exports, and energy demand. In the first phase, the preliminary performance was investigated of 19 different machine learning algorithms using 5-fold cross-validation and measured their performance using five different metrics: MSE, RMSE, MAE, R-squared, and MAPE. Secondly, ensemble models were constructed by utilizing individual machine learning algorithms, and the performance of these ensemble models was compared with both each other and the best-performing individual machine learning algorithm. The analysis of the results revealed that placing Ridge as the meta-learner and using ET, RF and Ridge as the base learners in the stacking ensemble model produced superior results compared to the other models across performance metrics. It is anticipated that the findings of this research can be applied globally and prove valuable for energy policy planning in any country. The results obtained not only highlight the accuracy and effectiveness of our predictive model but also underscore the broader implications of this study within the framework of the United Nations' Sustainable Development Goals (SDGs).

We investigated rainfall patterns in India over during the 60-year period from 1951 to 2010 and predicted changes for the next century (2051-2100) with an assumed 4K warming from large ensemble experiments (190 members). We focused on rainfall patterns during two periods of present-day climate (1951-1980 and 1981-2010) and their projected changes for the near and far future (2051-2080 and 2081-2110). Our analysis revealed that the northeastern region of India and some southern regions received higher rainfall during the period of 1951-2010. This finding is consistent with daily observations from the Asian Precipitation Highly Resolved Observational Data Integration Towards Evaluation (APHRODITE). In a warming climate, rainfall events in India are predicted to carry more precipitation, with the northeast and southern regions experiencing stronger rainfall events. The frequency and intensity of these events (with more than 20mm of rainfall per day, on average) are also expected to increase. Overall, our study suggests that water-related disasters such as flooding and landslides could be much worse in India in the future due to climate warming.

More than 110 countries including 500 cities worldwide have set the goal of reaching carbon neutrality. Heating contributes to most of the residential energy consumption and carbon emissions. The green energy transition of fossil-based heating systems is needed to reach the emission goals. However, the heating systems vary in energy source, heating technology, equipment location, and these complexities make it challenging for households to compare heating systems and make decisions. Hence, a decision support tool that provides a generalized ranking of individual heating alternatives is proposed for households as decision-makers to identify the optimal choice. This paper presents an analysis of 13 heating alternatives and 19 quantitative criteria in technological, environmental, and financial aspects, combines ideal solution based Multi-Criteria Decision Making with 6 weighting methods and 4 normalization methods, and introduces ensemble learning with a fuzzy membership function derived from Cauchy distribution to finalize the final ranking. The robustness of the proposed method is verified by 3 sensitive analyses from different aspects. Air to water heat pump, solar heating and direct district heating are the top three rankings in the final result under Danish national average data. A framework is designed to guide the decision-makers apply this ranking guideline with their practical feasible situations.

While the use of crime data has been widely advocated in the literature, its availability is often limited to large urban cities and isolated databases tend not to allow for spatial comparisons. This paper presents an efficient machine learning framework capable of predicting spatial crime occurrences, without using past crime as a predictor, and at a relatively high resolution: the U.S. Census Block Group level. The proposed framework is based on an in-depth multidisciplinary literature review allowing the selection of 188 best-fit crime predictors from socio-economic, demographic, spatial, and environmental data. Such data are published periodically for the entire United States. The selection of the appropriate predictive model was made through a comparative study of different machine learning families of algorithms, including generalized linear models, deep learning, and ensemble learning. The gradient boosting model was found to yield the most accurate predictions for violent crimes, property crimes, motor vehicle thefts, vandalism, and the total count of crimes. Extensive experiments on real-world datasets of crimes reported in 11 U.S. cities demonstrated that the proposed framework achieves an accuracy of 73 and 77% when predicting property crimes and violent crimes, respectively.

Day-ahead forecasting of electricity prices is important in deregulated electricity markets for all the stakeholders: energy wholesalers, traders, retailers, and consumers. Electricity price forecasting is an inherently difficult problem due to its special characteristic of dynamicity and non-stationarity. In this paper, we present a robust price forecasting mechanism that shows resilience towards aggregate demand response effect and provides highly accurate forecasted electricity prices to the stakeholders in a dynamic environment. We employ an ensemble prediction model in which a group of different algorithms participate in predicting the price for each hour of a day. We propose two different strategies, namely, Fixed Weight Method (FWM) and Varying Weight Method (VWM), for selecting each hour's expert algorithm from the set of participating algorithms. In addition, we utilize a carefully engineered set of features selected from a pool of features derived from information such as past electricity price data, weather data, and calendar data. The proposed ensemble model offers better results than both the Pattern Sequence-based Forecasting (PSF) method and our own previous work using Artificial Neural Networks (ANN) alone do on the datasets for New York, Australian, and Spanish electricity markets.

This paper explores the potential of using the SAM segmentator to enhance the segmentation capability of known methods. SAM is a promptable segmentation system that offers zero-shot generalization to unfamiliar objects and images, eliminating the need for additional training. The open-source nature of SAM on GitHub allows for easy access and implementation. In our experiments, we aim to improve the segmentation performance by providing SAM with checkpoints extracted from the masks produced by DeepLabv3+, then merging the segmentation masks provided by these two networks. Additionally, we examine the \enquote{oracle} method (as upper bound baseline performance), where segmentation masks are inferred only by SAM with checkpoints extracted from ground truth. In addition, we tested in the CAMO datasets an ensemble of PVTv2 transformers; combining the ensemble and SAM yields state-of-the-art performance in that dataset. The results of our study provide valuable insights into the potential of incorporating the SAM segmentator into existing segmentation techniques. We release with this paper the open-source implementation of our method.

Feature selection is a crucial step in machine learning, aiming to identify the most relevant features in high-dimensional data, in order to reduce the computational complexity of model development and improve its generalization performance. Ensemble feature ranking methods combine the results of several feature selection techniques to identify a subset of the most relevant features for a given task. In many cases, they produce a more comprehensive ranking of features than the individual methods used in them. This paper presents a novel approach to ensemble feature ranking, which uses a weighted average of the individual ranking scores calculated by the individual methods. The optimal weights are determined using a Taguchi-type design of experiments. The proposed methodology significantly improves classification performance on the CSE-CIC-IDS2018 dataset, particularly for attack types where traditional average-based feature ranking score combinations resulted in low classification metrics.

To effectively preserve marine environments and manage endangered species, it is necessary to employ efficient, precise, and scalable solutions for environmental monitoring. Ecoacoustics provides several benefits as it enables non-intrusive, prolonged sampling of environmental sounds, making it a promising tool for conducting biodiversity surveys. However, analyzing and interpreting acoustic data can be time-consuming and often demands substantial human supervision. This challenge can be addressed by harnessing contemporary methods for automated audio signal analysis, which have exhibited remarkable performance due to advancements in deep learning research. This paper introduces a research investigation into developing an automatic computerized system to detect dolphin whistles. The proposed method utilizes a fusion of various resnet50 networks integrated with data augmentation techniques. Through extensive experiments conducted on a publically available benchmark, our findings demonstrate that our ensemble yields significant performance enhancements across all evaluated metrics. The MATLAB/PyTorch source code is freely available at: https://github.com/LorisNanni/

Object classification is a crucial task in deep learning, which involves the identification and categorization of objects in images or videos. Although humans can easily recognize common objects, such as cars, animals, or plants, performing this task on a large scale can be time-consuming and error-prone. Therefore, automating this process using neural networks can save time and effort while achieving higher accuracy. Our study focuses on the classification step of human chromosome karyotyping, an important medical procedure that helps diagnose genetic disorders. Traditionally, this task is performed manually by expert cytologists, which is a time-consuming process that requires specialized medical skills. Therefore, automating it through deep learning can be immensely useful. To accomplish this, we implemented and adapted existing preprocessing and data augmentation techniques to prepare the chromosome images for classification. We used ResNet-50 convolutional neural networks and an ensemble approach to classify the chromosomes, obtaining state-of-the-art performance in the tested dataset.

Communication is a key method of expressing one's thoughts and opinions. Amongst many modalities, speech and writing are the most powerful and common forms of human communication. Analysing what and how people think has inherently been an interesting and progressive research domain. This includes bimodal sentiment analysis which is an emerging area in natural language processing (NLP) and has received a great deal of attention in recent years in a variety of areas including social opinion mining, health care, banking, and so on. At present, there are limited studies on bimodal conversational sentiment analysis as it proves to be a challenging area given the complex nature of the way humans express sentiment cues across various modalities. To address this gap, a comparison of the performance of multiple data modality models has been conducted on the MELD dataset, a widely-used dataset for benchmarking sentiment analysis within the research community. Our work then demonstrates the results of combining acoustic and linguistic representations. Lastly, our proposed neural network-based ensemble learning technique is employed over six transformer and deep learning-based models, achieving a State-Of-The-Art (SOTA) accuracy.

Covid-19 has caused infections and deaths worldwide. While research in the field of Data Science has contributed good predictions of positive Covid-19 case numbers, this study's review of literature shows there is little research in the use of variants of the virus in predictions. We set out to define and evaluate novel variant features. We find that features relating to variant trends, thresholds and amino acid substitutions are especially powerful in two tasks. In the first task, predicting Covid-19 case numbers, accuracy improved from 71.53% without variant features to 82.12% with variant features. In the second task, predicting transmission severity of variants between two classes, we created a method to build some variable ensembles through selecting appropriate models that are generated with variant features. The test results showed that our ensembles are more accurate and reliable. One particular ensemble of 14 models correctly classified 90.91% of variants, outperforming other models including the popular Random Forest ensemble. In addition, as the variant features have represented more underlying information about Covid-19 pathophysiology, our ensemble methods use only a few data samples to achieve an accurate prediction. The ensemble of 14 models uses only 50 cases of each variant, an ability that could be exploited for early detection of highly infectious variants. These research findings may benefit public health professionals, policy makers, and the research community in the collective efforts to overcome this disease.

Convolutional Neural Networks (CNNs) have gained prominence in the research literature on image classification over the last decade. One shortcoming of CNNs, however, is their lack of generalizability and tendency to overfit when presented with small training sets. Augmentation directly confronts this problem by generating new data points providing additional information. In this paper, we investigate the performance of more than ten different sets of data augmentation methods, with two novel approaches proposed here: one based on the Discrete Wavelet Transform and the other on the Constant-Q Gabor transform. Pretrained ResNet50 networks are finetuned on each augmentation method. Combinations of these networks are evaluated and compared across three benchmark data sets of images representing diverse problems and collected by instruments that capture information at different scales: a virus data set, a bark data set, and a LIGO glitches data set. Experiments demonstrate the superiority of this approach. The best ensemble proposed in this work achieves state-of-the-art performance across all three data sets. This result shows that varying data augmentation is a feasible way for building an ensemble of classifiers for image classification (code available at https://github.com/LorisNanni).

The novel coronavirus disease (COVID-19) has created immense threats to public health on various levels around the globe. The unpredictable outbreak of this disease and the pandemic situation are causing severe depression, anxiety and other mental as physical health related problems among the human beings. To combat against this disease, vaccination is essential as it will boost the immune system of human beings while being in the contact with the infected people. The vaccination process is thus necessary to confront the outbreak of COVID-19. This deadly disease has put social, economic condition of the entire world into an enormous challenge. The worldwide vaccination progress should be tracked to identify how fast the entire economic as well as social life will be stabilized. The monitor ofthe vaccination progress, a machine learning based Regressor model is approached in this study. This tracking process has been applied on the data starting from 14^th December, 2020 to 24^th April, 2021. A couple of ensemble based machine learning Regressor models such as Random Forest, Extra Trees, Gradient Boosting, AdaBoost and Extreme Gradient Boosting are implemented and their predictive performance are compared. The comparative study reveals that the AdaBoostRegressor outperforms with minimized mean absolute error (MAE) of 9.968 and root mean squared error (RMSE) of 11.133.

Recent studies on seafloor mapping have presented different modelling methods for the automatic classification of seafloor sediments. However, most of these studies have applied these models to seafloor data with appropriate number of ground-truth samples, which raises the question whether these methods are applicable to studies with smaller numbers of ground-truth data. In this study, we aim to address this issue by conducting sediment class-specific predictions using ensemble modelling to map areas with limited or without ground-truth data and combined with hydro-acoustic datasets. The resulting class-specific maps were then assembled into one map, where the most probable class was assigned to the appropriate location. Our approach was able to predict sediment classes without bias to the class with more ground-truth data and produced reliable seafloor sediment distributions maps that can be used for seafloor monitoring. Sediment shifts of a heterogenous seafloor in the Sylt Outer Reef, German North Sea were also assessed to understand the sediment dynamics in the area. The analyses of sediment shifts showed that the western area of the Sylt Outer Reef is highly active, and the results of the analyses assisted in providing recommendations on future seafloor monitoring activities.

In this study, we conducted experiments to assess the forecasting capabilities for the tropical cyclone (TC) genesis over the South China Sea using the ensemble-based data assimilation system (EPS-DA) by WRF-LETKF. These experiments covered forecast lead times of up to 5 days and spanned a period from 2012 to 2019, involving a total of 45 TC formation events. The evaluation involved forecast probability assessments and positional and timing error analysis. Results indicated that successful forecasting depends on lead time and initial condition quality. For TC formation from an embryo vortex to tropical depression intensity, the EPS-DA system demonstrated improved accuracy as the forecast cycle approached the actual formation time. TC centers converged toward observed locations, highlighting the potential of assimilation up to 5 days before formation. We examined statistical variations in dynamic and thermodynamic variables relevant to TC processes, offering an objective system assessment. Our study emphasized early warnings of TC development appear linked to formation-time environmental conditions, particularly strong vorticity and enhanced moisture processes.

In the present study, we implemented species distribution models (SDMs) to determine the potential invasive distribution range of a cichlid fish, Nile tilapia, Oreochromis niloticus, in a very sensitive inland water system in the southeastern Arabian Peninsula where it is now established and spreading. By using the “Biomod2” package, and the latest CMIP6 climate data, four single species distribution models and EM were built, and the results were evaluated by operating characteristic (ROC) and the true skill statistic (TSS). Based on the relative importance of climate factor analysis the mean diurnal range and precipitation seasonality were useful predictors of habitat suitability for O. niloticus. The response curve analysis indicated that the presence probability of O. niloticus decreased with increasing mean diurnal range and decreasing precipitation seasonality. The suitable distribution ranges for O. niloticus in the studied area were mainly distributed in the northeast of this region. The ensemble model results specified a significant impact of climate change on O. niloticus distribution, so highly suitable areas for this species will be reduced, while areas with low to moderate suitability increase slightly or remain unchanged. Based on the available data, Nile tilapia fulfills all four criteria to be considered as an invasive species: colonization in a new area, direct or indirect impacts of anthropological activity on its distribution range, geographical discontinuity between its native area (Africa) and the Arabian Peninsula (remote dispersal), and finally, its successful breeding in situ without human assistance resulting establishment of this biopollutant. Species distribution modeling provides insights into the factors that can reduce the risks and negative impacts of this biological pollutant on aquatic ecosystems of the southeastern Arabian Peninsula which harbors about 22 recognized fish species in 16 genera, 10 families, seven orders, and a class including 20 natives (seven endemic) species.

In the realm of computer vision, semantic segmentation is the task of recognizing objects in images at the pixel level. This is done by performing a classification of each pixel. The task is complex and requires sophisticated skills and knowledge about the context to identify objects’ boundaries. The importance of semantic segmentation in many domains is undisputed. In medical diagnostics, it simplifies the early detection of pathologies, thus mitigating the possible consequences. In this work, we provide a review of the literature on deep ensemble learning models for polyp segmentation and we develop new ensembles based on convolutional neural networks and transformers. The development of an effective ensemble entails ensuring diversity between its components. To this end, we combine different models (HarDNet-MSEG, Polyp-PVT, and HSNet) trained with different data augmentation techniques, optimization methods, and learning rates, which we experimentally demonstrate to be useful to form a better ensemble. Most importantly, we introduce a new method to obtain the segmentation mask which is more suitable for combining transformers in an ensemble. In our extensive experimental evaluation, the proposed ensembles exhibit state-of-the-art performance.

This study investigates the possible mechanisms related to the wide range of relative humidity in cirrus clouds (RHi). Under the closed adiabatic assumption, the impacts of vertical motion and ice crystal deposition/sublimation on RHi are investigated through in situ observations and parcel model simulations. Vertical motion is an active external force that changes the RHi, and ice crystal deposition/sublimation plays a role in mitigating the change in RHi. They are the two most important mechanisms involved in controlling the RHi fluctuation during cirrus evolution and could well explain the wide range of RHi in wave-related cirrus clouds. Furthermore, a comparison of cloud statistical characteristics from both observations and simulations shows that a very low value (e.g., 0.001) for the water vapor ice deposition coefficient is highly unlikely.

Climate change is expected to increase the occurrence of droughts with the hydrology in alpine systems being largely determined by snow dynamics. In this paper we propose a methodology to assess the impact of climate change on both meteorological and hydrological droughts taking into account the dynamics of the snow cover area (SCA). We will also analyse the correlation between these types of droughts. We have generated ensembles of local climate scenarios based on regional climate models (RCMs) representative of potential future conditions. We have considered several sources of uncertainty: different historical climate databases, simulations obtained with several RCMs, and some statistical downscaling techniques. We then used a stochastic weather generator (SWG) to generate multiple climatic series preserving the characteristics of the ensemble scenario. These were simulated within a cellular automata (CA) model to generate multiple SCA future series. They were used to calculate multiple series of meteorological drought indices, the Standardized Precipitation Index (SPI) and Standardized Precipitation Evapotranspiration Index (SPEI)) and a novel hydrological drought index (Standardized Snow Cover Index (SSCI)). A linear correlation analysis was applied to both types of drought to analyse how they propagate and the time delay between them. We applied the proposed methodology to the Sierra Nevada (southern Spain) where we estimated a general increase in meteorological and hydrological drought magnitude and duration for the horizon 2071-2100 under the RCP 8.5 emission scenario. The SCA droughts also revealed a significant increase in drought intensity. The meteorological drought propagation to SCA droughts is reflected in an immediate or short time (1 month), obtaining significant correlations in lower accumulation periods of drought indices (3 and 6 months). This will allow us to obtain information about meteorological drought from SCA deficits and vice versa.

Capability assessment plays a crucial role in the demonstration and construction of equipment. To improve the accuracy and stability of capability assessment, we study the neural network learning algorithms in the field of capability assessment and index sensitivity. Aiming at the problem of over-fitting and parameter optimization in neural network learning, the paper proposes an improved machine learning algorithm—the Ensemble Learning Based on Policy Optimization Neural Networks (ELPONN) algorithm with the policy optimization and ensemble learning. This algorithm presents optimized neural network learning algorithm through different strategies evolution, and builds an ensemble learning model of multi-intelligent algorithms to assessment the capability and analyze the sensitivity of the indexes. Through the assessment of capabilities, the algorithm effectively avoids parameter optimization from entering the minimum point in performance to improve the accuracy of equipment capability assessment, which is significantly better than previous neural network assessment methods. The experimental results show that the mean relative error is 4.10%, which is better than BP, GABP, and early stopping. The ELPONN algorithm has better accuracy and stability performance, and meets the requirements of capability assessment.

In this paper, we propose methods for brain tumor detection in MRI images based on ensemble learning. We build upon prior research on ensemble methods by testing the concatenation of pre-trained models: features extracted via transfer learning are merged and segmented by classification algorithms or a stacked ensemble of those algorithms. The proposed approach achieved accuracy scores of 0.98 , outperforming a benchmark VGG-16 model. Considerations to granular computing are given in the paper as well.

Melanoma is considered the most serious and aggressive type of skin cancer, and metastasis appears to be the most important factor in prognosticating this type of cancer. With the emergence of new therapeutic strategies for metastatic melanoma that have shown improvement in patient survival, we developed a transfer learning-based biomarker discovery model that could help in the diagnosis and prognosis of this disease. After applying it to the ensemble machine learning model, results reveal that the genes we found show consistency with other methodologies previously applied to the same TCGA (The Cancer Genome Atlas) data set, and our methods found novel biomarker genes as well. Our ensemble model achieved Area Under the Receiver Operating Characteristic (AUC) of 0.9861, an accuracy of 91.05, and an F1 score of 90.60 using an independent validation data set. This study was able to identify potential genes for diagnostic classification (C7 and GRIK5) and diagnostic and prognostic biomarkers (S100A7, S100A7, KRT14, KRT17, KRT6B, KRTDAP, SERPINB4, TSHR, PVRL4, WFDC5, IL20RB). We also assessed the potential sources of bias for our model and confirmed some of them by the model's performance.

Scattering hyperspectral technology is a nondestructive testing method with many advantages. Here, we propose a method to improve the accuracy of egg freshness, research the influence of incident angles of light source on the accuracy and explain its mechanism. A variety of weak classifiers classify eggs based on the spectra after preprocessing and feature wavelength extraction to obtain three classifiers with the highest accuracy. The three classifiers are used as metamodels of stacking ensemble learning to improve the highest accuracy from 96.25% to 100%. Moreover, the highest accuracy of scattering, reflection, transmission and mixed hyperspectral of eggs are 100.00%, 88.75%, 95.00% and 96.25%, respectively, indicating that the scattering hyperspectral for egg freshness detection is better than that of the others. In addition, the accuracy is inversely proportional to the angle of incidence due that the smaller the incident angle, the camera collects a larger proportion of scattering light, which contains more biochemical parameters of an egg than that of reflection and transmission. These results are very important for improving the accuracy of non-destructive testing and selecting the incident angle of the light source, and have potential applications in online non-destructive testing.

Due to the recent trend of unmanned economy, retail stores have gradually reduced their service and cashier manpower. The retail product recognition becomes one of key issues for unmanned shopping. Although the success of deep neural networks has made object recognition feasible in a variety of applications, it still struggles to perform well on a large number of classes of retail products. In this paper, we propose an improved ensemble learning-based approach for retail product recognition. In the proposed approach, the object classification network is first improved through feature extraction and block attention. An ensemble model is then built by integrating multiple network models and using loss selection as model weights. In the experiments, the feasibility of our ensemble learning-based approach has been evaluated through many production items. The results demonstrate the effectiveness of the proposed approach compared with previous retail product recognition methods.

To estimate the degree of quantum entanglement, it is important to understand the statistical behavior of functions of spectrum of density matrices such as von Neumann entropy, quantum purity, and entanglement capacity. These entangle- ment metrics over different generic state ensembles have been studied intensively in the literature. As an alternative metric, in this work we study sum of square root spectrum of density matrices, which is relevant to negativity and fidelity in quantum information processing. In particular, we derive the exact mean and vari- ance of sum of square root spectrum over the Bures-Hall generic state ensemble extending known results obtained recently over the Hilbert-Schmidt ensemble.

: Border management serves as a crucial control checkpoint for governments to regulate the quality and safety of imported food. In 2020, the ensemble learning prediction model EL V.1 was introduced to Taiwan's border food management. This model primarily assesses the risk of imported food by combining five algorithms to determine whether quality sampling should be performed on imported food at the border. In this study, a second-generation ensemble learning prediction model, EL V.2, was developed based on seven algorithms to enhance the "detection rate of unqualified cases" and improve the robustness of the model. The chi-square test was employed to compare the efficacy of "pre-launch (2019) random sampling inspection" and "post-launch (2020–2022) model prediction sampling inspection". For cases recommended for inspection by the ensemble learning model and subsequently inspected, the unqualified rates were 5.10%, 6.36%, and 4.39% in 2020, 2021, and 2022, respectively, which were significantly higher (p=0.000***) compared with the random sampling rate of 2.09% in 2019. The prediction indices established by the confusion matrix were used to further evaluate the prediction effects of EL V.1 and EL V.2, and the EL V.2 model exhibited superior predictive performance compared with EL V.1, and both models outperformed random sampling.

To improve the monitoring of the electrical power grid it is necessary to evaluate the influence of contamination in relation to leakage current and its progression to a disruptive discharge. In this paper, insulators were tested in a saline chamber to simulate the increase of salt contamination on their surface, and to evaluate the supportability of these components. From the time series forecasting of the leakage current, it is possible to evaluate the development of the fault before a flashover occurs. Choosing which method to use is always a difficult task since some models may have higher computational effort. In this paper, for a complete evaluation, the long short-term memory (LSTM), group method of data handling (GMDH), adaptive neuro-fuzzy inference system (ANFIS), bootstrap aggregation (bagging), sequential learning (boosting), random subspace, and stacked generalization (stacking) ensemble learning models are analyzed. A review and comparison of these well-established methods for time series forecasting is performed. From the results of the best structure of the model, the hyperparameters are evaluated and the Wavelet transform is used to obtain an enhanced model.

In this paper, we examine two strategies for boosting the performance of ensembles of Siamese networks (SNNs) for image classification using two loss functions (Triplet and Binary Cross Entropy) and two methods for building the dissimilarity spaces (FULLY and DEEPER). With FULLY, the distance between a pattern and a prototype is calculated by comparing two images using the fully connected layer of the Siamese network. With DEEPER, each pattern is described using a deeper layer combined with dimensionality reduction. The basic design of the SNNs takes advantage of supervised k-means clustering for building the dissimilarity spaces that train a set of support vector machines, which are then combined by sum rule for a final decision. The robustness and versatility of this approach are demonstrated on several cross-domain image data sets, including a portrait data set, two bioimage and two animal vocalization data sets. Results show that the strategies employed in this work to increase the performance of dissimilarity image classification using SNN is closing the gap with standalone CNNs. Moreover, when our best system is combined with an ensemble of CNNs, the resulting performance is superior to an ensemble of CNNs, demonstrating that our new strategy is extracting additional information.

This paper presents a comparison of the predictive capability of three hydrological models in a heavily influenced catchment in Peninsula India. In catchments where there is a high dependence on both streamflow and groundwater to meet demands, it is of importance to capture the catchment processes correctly. This study highlights the performance evaluation of a multi-model ensemble consisting of GWAVA (Global Water AVailability Assessment) model, SWAT (Soil Water Assessment Tool) and VIC (Variable Infiltration Capacity) model for comparative purposes and the key catchment hydrological processes. The three models were compared in several sub-catchments in the upstream reaches of the Cauvery river catchment. Model performances for monthly streamflow simulations from 1983 – 2005 were analysed for five catchments in the Upper Cauvery. The analysis was undertaken using Nash- Sutcliffe Efficiency, Kling- Gupta Efficiency and percent bias. Additionally, a mean ensemble is presented. The application of a multi-model ensemble approach can be useful in overcoming the uncertainties associated with individual models. The ensemble mean has a better predictive ability in catchments with reservoirs than the individual models. Utilising multiple models could be a suitable methodology to offset uncertainty in input data and poor reservoir operation functionality within individual models. This study has highlighted the importance of an accurate spatial representation of precipitation for input into hydrological models and comprehensive reservoir functionality is paramount to obtaining good results in this region.

There are uncertainties associated with every phase of the stock assessment process, ranging from the collection of data, assessment model choice, model assumptions and interpretation of risk to the implementation of management advice. The dynamics of fish populations are complex, and our incomplete understanding of those dynamics (and limited observations of important mechanisms) necessitate that models are simpler than nature. The aim is for the model to capture enough of the dynamics to accurately estimate trends and abundance and to provide advice to managers about sustainable harvests. The \textit{status quo} approach to assessment modelling has been to identify the `best' model, based on diagnostics and model selection criteria, and to generate advice from that model, mostly ignoring advice from other model configurations regardless of how closely they performed relative to the chosen model. We review the suitability of the ensemble modelling paradigm to more fully capture uncertainty in stock assessment model building and the provision of advice. We recommend further research to evaluate potential gains in modelling performance and advice from the use of ensemble modelling, while also suggesting revisions to the formal process for reviewing models and providing advice to management bodies.

Lower Back Pain (LBP) is a disease that needs immediate attention. Person with back pain shall go immediately to doctor for treatment. Injury, excessive works and some medical conditions are result of back pain. Back pain is common to any age of human for different reasons. Due to factors such as previous occupation and degenerative disk disease the chance of developing lower back pain increases for older people. It hampers the working condition of people common reason for seeking medical treatment. The result is absence from work and is unable to normal due to pain. It creates uncomfortable and debilitating situations. Hence, detecting this disease at an early stage will assist the medical field experts to suggest counter measures to the patients. Detection of lower back pain is implemented in this paper by applying ensemble machine learning technique. This paper proposes Stacking ensemble classifier as an automated tool that will predict lower back pain tendency of a patient. Experimental result implies that the proposed method reaches an accuracy of 76.34%, f1-score of 0.76 and MSE of 0.34.

This study proposes a novel hybrid machine learning approach for modeling of rainfall-induced shallow landslides. The proposed approach is a combination of an instance-based learning algorithm (k-NN) and Rotation Forest (RF), state of the art machine techniques that have seldom explored for landslide modeling. The Lang Son city area (Vietnam) is selected as a case study. For this purpose, a spatial database for the study area was constructed, and then, was used to build and evaluate the hybrid model. Performance of the model was assessed using Receiver Operating Characteristic (ROC), area under the ROC curve (AUC), success rate and prediction rate, and several statistical evaluation metrics. The results showed that the model has high performance with both the training data (AUC = 0.948) and the validation data (AUC = 0.848). The results were compared with those obtained from soft computing techniques i.e. Random Forest, J48 Decision Trees, and Multilayer Perceptron Neural Networks. Overall, the performance of the proposed model is better than those obtained from the above methods. Therefore, the proposed model is a promising tool for landslide modeling. The research result can be highly useful for land use planning and management in landslide prone areas.

The lack of a generalizable machine learning model for predicting the safety of food for 1 human consumption is a significant challenge for policymakers and responsible authorities. This 2 study provides a step-by-step guide to predict the results of seafood product import inspections, 3 focusing on identifying and understanding the critical factors that influence these results. By compar- 4 ing the performances of an ensemble of machine learning models, this study combines the strengths 5 of multiple algorithms to improve the predictive accuracy and gain insights into the key factors 6 impacting them. The ensemble model based on the soft voting technique achieves superior perfor- 7 mance to that based on the hard voting technique in terms of the recall and area under the curve 8 (AUC) scores. The study discovered that various characteristics, such as the exporting country ratio, 9 major product category, overseas manufacturer ratio, importer ratio, and seasonal variation, had a 10 substantial influence on the models’ decisions. This research guide for predicting seafood product 11 import inspection results could pave the path for other items to follow.

Ensemble fuzzing in parallel with heterogeneous fuzzers has been proposed to leverage the advantages of diverse fuzzers and improve testing efficiency. However, in current ensemble fuzzing methods, the collaboration among different fuzzers is achieved solely by synchronizing the seeds discovered by each fuzzer. This results in a high likelihood of different fuzzers choosing the same seeds and creating a large number of equivalent testcases, thus reducing overall fuzzing efficiency. Meanwhile, the existing task division method proposed by AFLTeam is highly coupled with the fuzzer specially designed for it, making it challenging to apply to ensemble fuzzing directly. So, in this paper, we proposed a callgraph-based task division method suitable for ensemble fuzzing. Firstly, we divided the target program’s callgraph into subgraphs (subtasks) balancing expected workloads. Then, we divided the global seed corpus into sub-corpora, each corresponding to a subtask, making fuzzers easily accept the subtasks. Finally, we designed synchronization mechanisms for coverage bitmaps and seeds to realize the collaborative fuzzing among different fuzzers, and a cyclic subtask scheduling strategy to fully leverage the benefits of ensemble fuzzing. We have implemented a prototype called TAEF. The evaluation results show that in the best-case scenario, our method has up to 24% more branch coverage than previous work.

Large terrestrial carnivores play a crucial role in top-down control within terrestrial ecosystems, maintaining ecosystem stability and biodiversity. However, intense interspecific competition often arises among sympatric large carnivores, leading to population reductions or even extinctions. Spatial partitioning through divergent habitat selection helps mitigate such competition. In Xinlong County, Sichuan Province, we used 293 infrared cameras for monitoring from September to May 2016 and March to October 2022. By employing the Generalized Linear Model (GLM) and the Maximum Entropy Model (MaxEnt), we developed an ensemble model predicting the suitable habitat distribution of leopards (Panthera pardus) and wolves (Canis lupus). We analyzed the main environmental factors influencing habitat selection and the fragmentation of suitable habitats. We found that suitable habitat distribution differed significantly between them. Both species preferred areas with gentle slopes close to settlements. While leopards' habitat selection primarily depended on the distance from settlements, the slope was predominant for wolves. Suitable habitats displayed aggregation, yet wolves exhibited higher fragmentation and more complex patch shapes, indicating greater susceptibility to human activities. These results suggest that sympatric large carnivores, such as leopards and wolves, can reduce spatial competition intensity and promote spatial partitioning by selecting divergent suitable habitats, thereby facilitating species coexistence.

When integrating data from different sources, there are problems of synonymy, different languages, concepts of different granularity. This paper proposes a simple but effective approach to evaluate the semantic similarity of short texts, especially keywords. The method is capable of matching keywords from different sources and languages by exploiting transformers and WordNet-based methods. Key features of the approach include its unsupervised pipeline, mitigation of the lack of context in keywords, scalability for large archives, support for multiple languages and real-world scenarios adaptation capabilities. The work aims to provide a versatile tool for different cultural heritage archives without requiring complex customization. The objectives of the paper are to explore different approaches to identifying similarities in 1- or n-gram tags, to evaluate and compare different pre-trained language models, and to define integrated methods to overcome limitations. Tests to validate the approach have been conducted using the QueryLab portal, a search engine for cultural heritage archives, to evaluate the proposed pipeline.

InSAR and associated analytic methods enable relative surface deformation measurements from low Earth orbit with a potential accuracy of centimeters to millimeters. However, assessing the actual accuracy can be quite difficult. The analytic methods are complicated enough that naive analytic error propagation is infeasible, and, in many settings, InSAR practitioners lack sufficient ground truth to assess results. Phase noise due to partial decorrelation from changes in the scattering properties of the ground is a prominent source of accuracy loss. In this paper we present a method to assess the loss of precision due to this component of phase noise. The proposed method consists of generating synthetic data stacks whose statistical properties match those of the actual input SAR data stacks, and then using the synthetic data for an ensemble calculation. The spread of the results of the ensemble calculation indicates the loss of precision. We show examples of the ensemble analysis at a mining operation in South Africa, and demonstrate the ability to estimate the precision of two InSAR deformation retrieval methods on a point-by-point and epoch-by-epoch basis.

Ultrasound (US) is commonly used for the diagnosis of liver masses. Ensemble learning has been widely used for image classification, but its methods have not been fully optimized. This study was performed to investigate the usefulness of ensemble learning and compare a number of ensemble learning techniques using multiple convolutional neural network (CNN)-trained models for image classification of liver masses in US images. The US imaging data set was classified into four categories: benign liver tumor (BLT, 6320 images), liver cyst (LCY, 2320 images), metastatic liver cancer (MLC, 9720 images), and primary liver cancer (PLC, 7840 images). In this study, 250 test images were randomly selected for each class, for a total of 1000 images, and the remaining images were used for training. Sixteen different CNNs were used for to train and test the US images. All four types of ensemble learning—soft voting (SV), stacking (ST), weighted average voting (WAV), and weighted hard voting (WHV)—showed greater accuracy than the single CNN. All four types also showed significantly better deep learning performance than ResNeXt101 alone. For image classification of liver masses using US images, ensemble learning improved the performance of deep learning over a single CNN.

Ransomware is one of the most dangerous types of malware, which is frequently intended to spread through a network to damage the designated client by encrypting the client’s vulnerable data. Conventional signature-based ransomware detection technique falls behind because it can only detect known anomalies. When it comes to new and non-familiar ransomware traditional system unveils huge shortcomings. For detecting unknown patterns and sorts of new ransomware families,behavior-based anomaly detection approaches are likely to be the most efficient approach. In the wake of this alarming condition, this paper presents an ensemble classification model consisting of three widely used machine learning techniques that include Decision Tree (DT), RandomForest (RF), and K-nearest neighbor (KNN). To achieve the best outcome ensemble soft voting and hard voting techniques are used while classifying ransomware families based on attack attributes. Performance analysis is done by comparing our proposed ensemble models with standalone models on behavioral attributes based ransomware dataset..

Tephra plumes can cause a significant hazard for surrounding towns, infrastructure, and air traffic. The current work presents the use of a small and compact X-band Multi-Parameter (X-MP) radar for the remote tephra detection and tracking of two eruptive events at Merapi Volcano, Indonesia, in May and June 2018. Tephra detection was done by analysing the multiple parameters of radar: copolar correlation and reflectivity intensity. These parameters were used to cancel unwanted clutter and retrieve tephra properties, which are grain size and concentration. Real-time spatial and temporal forecasting of tephra dispersal was performed by applying an advection scheme (nowcasting) in the manner of Ensemble Prediction System (EPS). Cross-validation was done using field-survey data, radar observations, and Himawari-8 imagery. The nowcasting model computed both the displacement and growth and decaying rate of the plume based on the temporal changes in two-dimensional movement and tephra concentration, respectively. Our results with ground-based data, where the radar-based estimated grain size distribution fell within the range of in-situ data. The uncertainty of real-time forecasted tephra plume depends on the initial condition, which affects the growth-and decaying rate estimation. The EPS improves the predictability rate by reducing the number of missed and false forecasted events. Our findings and the method presented here are suitable for early warning of tephra fall hazard at the local scale.

This paper provides the state of the art of data science in economics. Through a novel taxonomy of applications and methods advances in data science are investigated. The data science advances are investigated in three individual classes of deep learning models, ensemble models, and hybrid models. Application domains include stock market, marketing, E-commerce, corporate banking, and cryptocurrency. Prisma method, a systematic literature review methodology is used to ensure the quality of the survey. The findings revealed that the trends are on advancement of hybrid models as more than 51% of the reviewed articles applied hybrid model. On the other hand, it is found that based on the RMSE accuracy metric, hybrid models had higher prediction accuracy than other algorithms. While it is expected the trends go toward the advancements of deep learning models.

Brain Tumor is one of the severe diseases and occurrence of this disease threats human life. Detection of brain tumor in advance can secure patient’s life from unwanted loss. Well-timed and swift disease detection and treatment strategy can lead to improved quality of life in these patients. This paper attempts to use Machine Learning based ensemble approaches for recognising patients with brain tumor. Ensemble technique based AdaBoost classifier and 10-fold stratified cross-validation method are assembled in single platform is proposed in this paper for prediction of brain tumor. This prediction is compared against three baseline classifiers such as Gradient Boost, Random Forest and Extra Trees classifier. Experimental result implies the superiority of this model with an accuracy of 98.97%, f1-score of 0.99, kappa statistics score of 0.95 and MSE of 0.0103.

Nowadays, we live in a world in which people are facing with a lot of data that should be stored or displayed. One of the key methods to control and manage this data refers to grouping and classifying them in clusters. Today, clustering has a critical role in information retrieval methods for organizing large collections inside a few significant clusters. One of the main motivations for the use of clustering is to determine and reveal the hidden and inherent structure of a set of data. Ensemble clustering algorithms combine multiple clustering algorithms to finally reach an overall clustering system. Ensemble clustering methods by lack of information fusing utilize several primary partitions of data to find better ways. Since various clustering algorithms look at the different data points, they can produce various partitions from such data. It is possible to create a partition with high performance by combining the partitions obtained from different algorithms, even if the clusters to be very dense from each other. Most studies in this area have examined all the initial clusters. In this study, a new method is used in which the most sustainable clusters are utilized instead of all primary produced clusters. Consensus function based on co-association matrixes used to select more stable clusters. The most stable clusters selection method is done by cluster stability criterion based on F-measure. Optimization functions are used to optimize the obtained final clusters. The genetic algorithm is the optimizer used in this article to find the ultimate clusters participated in a consensus. Experimental results on several datasets show that the output of proposed method is various clusters with high stability.

We investigate the stationary and dynamic properties of the celebrated Nosé-Hoover dynamics of many-body interacting Hamiltonian systems, with an emphasis on the effect of inter-particle interactions. To this end, we consider a model system with both short- and long-range interactions. The Nosé-Hoover dynamics aims to generate the canonical equilibrium distribution of a system at a desired temperature by employing a set of time-reversible, deterministic equations of motion. A signature of canonical equilibrium is a single-particle momentum distribution that is Gaussian. We find that the equilibrium properties of the system within the Nosé-Hoover dynamics coincides with that within the canonical ensemble. Moreover, starting from out-of-equilibrium initial conditions, the average kinetic energy of the system relaxes to its target value over a size-independent timescale. However, quite surprisingly, our results indicate that under the same conditions and with only long-range interactions present in the system, the momentum distribution relaxes to its Gaussian form in equilibrium over a scale that diverges with the system size. On adding short-range interactions, the relaxation is found to occur over a timescale that has a much weaker dependence on system size. This system-size dependence of the timescale vanishes when only short-range interactions are present in the system. An implication of such an ultra-slow relaxation when only long-range interactions are present in the system is that macroscopic observables other than the average kinetic energy when estimated in the Nosé-Hoover dynamics may take an unusually long time to relax to its canonical equilibrium value. Our work underlines the crucial role that interactions play in deciding the equivalence between Nosé-Hoover and canonical equilibrium.

Accurate and efficient medicinal plant image classification is of utmost importance as these plants produce a wide variety of bioactive compounds that offer therapeutic benefits. With a long history of medicinal plant usage, different parts of plants, such as flowers, leaves, and roots, have been recognized for their medicinal properties and are used for plant identification. However, leaf images are extensively used due to their convenient accessibility and are a major source of information. In recent years, transfer learning and fine-tuning, which use pre-trained deep convolutional networks to extract pertinent features, has emerged as an extremely effective approach for image identification problems. This study is leveraging the power of three component deep convolutional neural networks, namely VGG16, VGG19 and DenseNet201, to derive features from the input images of the medicinal plant dataset, containing leaf images of 30 classes. The models were compared and ensembled to make four hybrid models to enhance the predictive performance by utilizing the averaging and weighted averaging strategies. Quantitative experiments were carried out to evaluate the models on the Mendeley medicinal leaf dataset. The resultant ensemble of VGG19+DensNet201 with fine-tuning showcases enhanced capability in identifying medicinal plant images with an improvement of 7.43% and 5.8% compared with VGG19 and VGG16. Furthermore, VGG19+DensNet201 can outperform its standalone counterparts by achieving an accuracy of 99.12% on the test set. A thorough assessment with metrics such as accuracy, recall, precision, and f1-score firmly establishes the effectiveness of the ensemble strategy.

Using traditional methods based on detection rules written by human security experts, there is a significant challenge in accurately detecting many network threats that are increasingly becoming sophisticated. In order to deal with the limitation of the traditional methods, network threat detection techniques utilizing artificial intelligence technologies such as machine learning are being extensively studied. Research has also been conducted on analyzing various string patterns in network packet payloads through natural language processing techniques to detect attack intent. However, due to the nature of packet payloads containing binary data as well as text data, a new approach is needed that goes beyond typical natural language processing techniques. In this paper, we study a token extraction method optimized for payloads using n-gram and byte pair encoding techniques. Furthermore, we generate embedding vectors that can understand the context of the packet payload using algorithms such as Word2Vec and FastText. We also compute the embedding of various header data associated with packets, such as IP addresses and ports. Given the combination of these features, we ensemble a text 1D-CNN and a multi-head attention network in a novel fashion. We validated the effectiveness of our classification technique with the CICIDS2017 open dataset and over half a million data collected by The Education Cyber Security Center (ECSC) currently operating in South Korea. The proposed model showed remarkable progress compared to previous studies, demonstrating a highly accurate classification performance with an F1-Score of 0.998. Our model can also preprocess and classify 150,000 network threats per minute to help security agents in the field divert their time to analyzing more complex attack patterns.

Ship pilots and maritime safety administration have an urgent need for more accurate and earlier warnings on strong wind gusts. This study firstly investigates the “Oriental Star” cruise ship cap-sizing event in 2015, one of the deadliest shipwreck events in recent years, and explores all related hydro-meteorological components in a global mesoscale model. It is found that, rather than the missing signal in raw surface wind prediction, the cumulus precipitation variable (CP) increases dramatically during the accident occurrence, which significantly corresponds to the sub-grid strong wind gust. The effective lead-time can be extended from 24hr (deterministic model) to 48hr (en-semble model). This finding is then verified in another two recent deadly cruise boat accidents. The introduction of the new variable is bearing the hope of improvement to the current maritime safeguard system, on predicting sub-grid strong wind gusts for small-size cruise boats in offshore and inland rivers. Finally, an automatic-response system is developed to provide economical con-vection prediction via Inmarsat email communication, aiming to explore operational severe con-vective gust early-warning and appropriate numerical mesoscale model application.

This study aims to determine the influence of selected social and health factors on the quality of inpatient treatments when regional health organizations use stacking ensemble model. The proposed procedure is based on the logistic regression method, which is used for direct prediction in the case of good fitting data and impossibility of including aggregation of classification algorithms but also in the opposite case, for fine calibration to obtain prediction. In opposite case, the procedure uses classification algorithms and several filter methods which initially rank individual factors according to their importance to reduce the dimensionality of problem and, in such way. obtaining one optimized classification prediction. The proposed procedure is trained using one part and tested using another part of dataset from case study which enables the generalization of the solution to the set goal and is verified by the Kaggle dataset Brest cancer. Case study was conducted using the real data acquired in the region connected to the city of Nis, Republic of Serbia. The obtained results show that the proposed model achieves better results that each of methods included in this stacking ensemble-regression and classification used individually.

This paper presents a study of an automated system for identifying planktic foraminifera at the species level. The system uses a combination of deep learning methods, specifically Convolutional Neural Networks (CNNs), to analyze digital images of foraminifera taken at different illumination angles. The dataset is composed of 1437 groups of sixteen grayscale images, one group for each foraminifer, that are then converted to RGB images with various processing methods. These RGB images are fed into a set of CNNs, organized in an Ensemble Learning (EL) environment. The ensemble is built by training different networks using different approaches for creating the RGB images. The study finds that an ensemble of CNN models trained on different RGB images improves the system's performance compared to other state-of-the-art approaches. The proposed system was also found to outperform human experts in classification accuracy.

For the soil in different regions, the nutrient fertility contained in it is different, and the detection and zoning management of soil nutrients before tillage every year can improve grain yield. In this paper, an integrated learning strategy model based on black soil hyperspectral data is designed for rapid classification of organic matter content classification of black soil. Soil hyperspectral image dataset of Xiangyang Experimental Base was collected; by changing the internal structure of the stacking model, an LSVM-stacking model with (MLP, SVC, DTree, XGBl, kNN) five classifiers as the L1 layer was built, and the simulated annealing algorithm was used for hyperparameter optimization. Compared to other stacking models, the LSVM-stacking metrics are significantly improved. The accuracy rate of hyperparameter optimization is improved by 38.6515%, the accuracy rate of the independent test data set is 0.9488, and the comparison of individual learners can improve the recognition classification accuracy of label"1" to 1.0.

An electrocardiograph (ECG) reflects the health of the human heart and is used to help diagnose arrhythmia and myocardial infarction(MI) in clinical practice. Early diagnosis of arrhythmia helps implement preventive measures and plays a crucial role in saving a patient's life. With the increasing demand of clinicians for ECG analysis technology, intelligent detection and diagnosis of ECG signals has become a more efficient means to assist physicians in diagnosing cardiovascular diseases. This paper introduces an ECG diagnosis approach based on an ensemble deep learning combination of CNN(convolutional neural network) and SLAP(stacked-long short term memory architecture for prediction) architecture. ECG data is denoised and further divided into single heartbeats to achieve data standardization and sample diversity. Adam optimizer and BCEwithlogitsloss multi-classification loss function were used to enhance the model effect, and the system achieved the classification effect of 99.3% average accuracy, 99.0% F1-value, and 99.2% sensitivity in MIT-BIH standard database classification. It also shows good generalization ability on the Tianchi data set.

In recent years, the field of deep learning achieved considerable success in pattern recognition, image segmentation and may other classification fields. There are a lot of studies and practical applications of deep learning on images, video or text classification. In this study, we suggest a method for changing the architecture of the most performing CNN models with the aim of designing new models to be used as stand-alone networks or as a component of an ensemble. We propose to replace each activation layer of a CNN (usually a ReLu layer) by a different activation function stochastically drawn from a set of activation functions: in this way the resulting CNN has a different set of activation function layers.

This study has simulated the typical rainstorm on 20 July 2021 over central east China by using the first-generation Chinese Reanalysis datasets and Global Land Data Assimilation System datasets, and the Noah land surface model coupled with the advanced weather research and forecasting model. Based on this, the gridded planetary boundary layer (PBL) profiles and ensemble states within soil perturbations are collected to investigate the main land-atmosphere coupling characteristics during this modeled rainstorm by using various local coupling metrics and the introduced ensemble statistical metrics. Results have shown that (1) except for the stratospheric thermodynamics and surface thermal over mountain areas, the main characteristics of mid-low layers and surface have been well documented in this modeled rainstorm; (2) the typical coupling intensity is characterized by the dominant morning moistening, early noon weak PBL warming around 2, noontime buoyant mixing temperature deficit around 274 K, daytime PBL and surface latent flux contribution around 100 and 280 W/m2 respectively, and significant afternoon soil-surface latent flux coupling; (3) moist static energy is more significant than PBL height during the relation chains, which is consistent with the significance of surface moistening indicated by local coupling metrics. In general, wet soil contributes greatly to daytime moisture evaporation, which then increases the early noon PBL warming and enhances the noontime buoyant mixing within weak flux contribution. However, this has been suppressed by large-scale forcing such as the upper southwestern inflows of rainstorms, which has further significantly shaped the spatial distribution of statistical metrics in contrast. These quantitatively described local couplings have highlighted both the convection potential diagnoses usage for the local weather application and more applicable coupling threshold diagnoses within the finer spatial investigation.

The Internet of Things technology opens the horizon for a broader scope of intelligent applications in smart cities. However, the massive amount of traffic exchanged among devices may cause security risks, significantly when devices are compromised or vulnerable to cyber-attack. An intrusion detection system is the most powerful tool to detect unauthorized attempts to access smart systems. It identifies malicious and benign traffic by analyzing network traffic. In most cases, only a fraction of network traffic can be considered malicious. As a result, it is difficult for an intrusion detection system to detect attacks at high detection rates while maintaining a low false alarm rate. This work proposes an integrated framework to detect suspicious traffic to address secure data communication in smart cities. This paper presents an approach to developing an intrusion detection system to detect various attack types. It can be done by implementing a Principal Component Analysis method that eliminates redundancy and reduces system dimensionality. Furthermore, the proposed model shows how to improve intrusion detection system performance by implementing an ensemble model.

Chemical elements in space can be synthesized by stellar nuclear reactors. Studying the dynamics of processes occurring in the stars introduces a concept of the ensemble-averaged stellar reactor. For future interstellar missions, the terrestrial and solar abundances were compared with considerable number of stars allocated in the ~ 200 pc solar neighborhood. According to the value of the effective temperature, four stellar classes are distinguished, for which the correlation coefficients and standard deviation are calculated. The statement about the possibility of transferring heavy elements synthesized by stars over long distances in space has been completely refuted. There is no immutability of the distribution of elements on neighboring stars and in the Solar System. It is shown that chemical elements are mainly synthesized inside each stellar reactor. The theory of the buoyancy of elements is generalized to stars. It has been suggested that stars overheat due to a shift in the parameters of nuclear processes occurring inside stars, which leads to the synthesis of heavy elements.

Real-world optimization problems are often governed by one or more constraints. Over the last few decades, extensive research has been performed in Constrained Optimization Problems (COPs) fueled by advances in computational intelligence. In particular, Evolutionary Algorithms (EAs) are a preferred tool for practitioners for solving these COPs within practicable time limits. We propose an ensemble of multi- method hybrid EA framework with four mutation operators, two crossover operators, multi-search [Differential Evolution (DE) & Gaining Sharing Knowledge (GSK)] optimization algorithm, and ensemble of constraint handling techniques to solve global real- world constrained optimization problem. The proposed frame- work FEPEA has an ascendancy of multiple adaptation strategies concerning the control parameters, search mechanisms, two sub-populations as well as uses knowledge sharing mechanism between junior and senior phases. The algorithm also combines the power of four popular constraint handling techniques (CHT) and uses a voting mechanism to select any particular CHT. On top of that, this algorithm also uses both linear and non- linear population size reduction in every step of the evolutionary process. We test our method on 57 real-world problems provided as part of the CEC 2020 special session & competition on real- world constrained optimization benchmark suite. Experimental results indicate that FEPEA is able to achieve state-of-the- art performance on real-world constrained global optimization when compared against other well-known real-world constrained optimizers.

The classifier system proposed in this work combines the dissimilarity spaces produced by a set of Siamese neural networks (SNNs) designed using 4 different backbones, with different clustering techniques for training SVMs for automated animal audio classification. The system is evaluated on two animal audio datasets: one for cat and another for bird vocalizations. Different clustering methods reduce the spectrograms in the dataset to a set of centroids that generate (in both a supervised and unsupervised fashion) the dissimilarity space through the Siamese networks. In addition to feeding the SNNs with spectrograms, additional experiments process the spectrograms using the Heterogeneous Auto-Similarities of Characteristics. Once the similarity spaces are computed, a vector space representation of each pattern is generated that is then trained on a Support Vector Machine (SVM) to classify a spectrogram by its dissimilarity vector. Results demonstrate that the proposed approach performs competitively (without ad-hoc optimization of the clustering methods) on both animal vocalization datasets. To further demonstrate the power of the proposed system, the best stand-alone approach is also evaluated on the challenging Dataset for Environmental Sound Classification (ESC50) dataset. The MATLAB code used in this study is available at https://github.com/LorisNanni.

This paper introduces a novel ensemble adjustment Kalman filter (EAKF) that integrates a machine-learning approach. The approach employs nonlinear variable relationships established by a deep neural network (DNN) during the analysis stage of the EAKF. This process nonlinearly projects observation increments into the state variable space. The newly developed DNN-EAKF algorithm can be applied to coupled data assimilation using coupled ocean-atmosphere models. It enhances cross-component updates in strongly coupled data assimilation (SCDA) by diminishing errors in estimating cross-component error covariance arising from insufficient ensemble members, thereby improving the SCDA analysis. This paper employs a conceptual model to conduct twin experiments, validating the DNN-EAKF’s capability to utilize cross-component observation information in SCDA effectively. The approach is anticipated to offer insights for future methodological integrations of machine learning and data assimilation and provide methods for SCDA applications in coupled general circulation models.

Due to the intermittency and fluctuation of photovoltaic (PV) output power, a high proportion of grid-connected PV power generation systems has a significant impact on power systems. Accurate PV power forecasting can alleviate the uncertainty of the PV power and is of great significance for the stable operation and scheduling of the power systems. Therefore, in this study, a feature rise-dimensional (FRD) two-layer ensemble learning (TLEL) model for short-term PV power deterministic forecasting and probability forecasting is proposed. First， based on the eXtreme Gradient Boosting (XGBoost), Random Forest (RF), CatBoost, and Long-short-term memory (LSTM) models, a TLEL model is constructed utilizing the ensemble learning algorithm. Meanwhile, the FRD method is introduced to construct the FRD-XGBoost-LSTM (R-XGBL), FRD-RF-LSTM(R-RFL), and FRD- CatBoost - LSTM (R-CatBL) models. Subsequently, the above models are combined to construct the FRD-TLEL model for deterministic forecasting, and perform probability interval forecasting based on quantile regression(QR). Finally, the performance of the proposed model is demonstrated with a real-world dataset. By comparing with other models, the proposed model displays better forecasting accuracy for deterministic forecasting and reliable forecasting intervals for probability forecasting, and good generalization ability in the datasets of different seasons and weather types.

Protein phosphorylation and ADP-ribosylation (ADPr), as two types of post-translational modifications (PTM), are the process of adding phosphate group and ADP-ribose moieties to proteins, respectively. Although both PTM types can occur on many amino acid types, serine is the most common. Serine phosphorylation (pS), serine ADPr (SADPr), and their in situ crosstalks (pSADPr) play essential roles in biological processes. Although in silico classifiers have been developed for predicting pS and SADPr sites, the classifier for predicting pSADPr sites is unavailable. In this study, we developed classifiers to predict pSADPr sites. Specifically, we collected 3250 human pSADPr, 7520 SADPr, 151,227 pS and 80,096 unmodified serine sites. Based on them, we investigated the characteristics of pSADPr sites and constructed three classifiers to predict pSADPr sites from the pS dataset, the SADPr dataset and the protein sequences separately. We built and evaluated five deep-learning classifiers in ten-fold cross-validation and independent test datasets. Three of them (e.g. Convolutional Neural Network with the One-Hot encoding, dubbed CNNOH) performed better than the rest two. For instance, CNNOH had the AUC values of 0.700, 0.914 and 0.954 for recognizing pSADPr sites from the SADPr, pS and unmodified serine sites.Therefore, it is challenging to distinguish pSADPr sites from SADPr sites compared to the other two. It is consistent with our observation that pSADPr's characteristics are more similar to those of SADPr than the rest. Furthermore, we used the classifiers as base classifiers to develop a few stacking-based ensemble classifiers to improve performance. However, none of the ensemble classifiers showed better performances, suggesting that the base classifiers have good enough performances. Finally, we developed an online tool for extensively predicting human pSADPr sites based on the CNNOH classifier, dubbed EdeepSADPr. It is freely available through http://edeepsadpr.bioinfogo.org/.

Although automated Acute Lymphoblastic Leukemia (ALL) detection is essential, it is challenging due to the morphological correlation between malignant and normal cells. The traditional ALL classification strategy is arduous, time-consuming, often suffers inter-observer variations, and necessitates experienced pathologists. This article has automated the ALL detection task, employing deep Convolutional Neural Networks (CNNs). We explore the weighted ensemble of deep CNNs to recommend a better ALL cell classifier. The weights are estimated from ensemble candidates' corresponding metrics, such as accuracy, F1-score, AUC, and kappa values. Various data augmentations and pre-processing are incorporated for achieving a better generalization of the network. We train and evaluate the proposed model utilizing the publicly available C-NMC-2019 ALL dataset. Our proposed weighted ensemble model has outputted a weighted F1-score of 88.6%, a balanced accuracy of 86.2%, and an AUC of 0.941 in the preliminary test set. The qualitative results displaying the gradient class activation maps confirm that the introduced model has a concentrated learned region. In contrast, the ensemble candidate models, such as Xception, VGG-16, DenseNet-121, MobileNet, and InceptionResNet-V2, separately produce coarse and scatter learned areas for most example cases. Since the proposed ensemble yields a better result for the aimed task, it can experiment in other domains of medical diagnostic applications.

One of the most basic pieces of information gained from dynamic electromyography is accurately defining muscle action and phase timing within the gait cycle. The human gait relies on selective timing and the intensity of appropriate muscle activations for stability, loading, and progression over the supporting foot during stance, and further to advance the limb in the swing phase. A common clinical practice is utilizing a low-pass filter to denoise integrated electromyogram (EMG) signals and to determine onset and cessation events using a predefined threshold. However, the accuracy of the defining period of significant muscle activations via EMG varies with the temporal shift involved in filtering the signals; thus, the low-pass filtering method with a fixed order and cut-off frequency will introduce a time delay depending on the frequency of the signal. In order to precisely identify muscle activation and to determine the onset and cessation times of the muscles, we have explored here onset and cessation epochs with denoised EMG signals using different filter banks: the wavelet method, empirical mode decomposition (EMD) method, and ensemble empirical mode decomposition (EEMD) method. In this study, gastrocnemius muscle onset and cessation were determined in sixteen participants within two different age groups and under two different walking conditions. Low-pass filtering of integrated EMG (iEMG) signals resulted in premature onset (28% stance duration) in younger and delayed onset (38% stance duration) in older participants, showing the time-delay problem involved in this filtering method. Comparatively, the wavelet denoising approach detected onset for normal walking events most precisely, whereas the EEMD method showed the smallest onset deviation. In addition, EEMD denoised signals could further detect pre-activation onsets during a fast walking condition. A comprehensive comparison is discussed on denoising EMG signals using EMD, EEMD, and wavelet denoising in order to accurately define an onset of muscle under different walking conditions.

Users of meteorological forecasts are often faced with the question of whether to make a decision now based on the current forecast or whether to wait for the next and hopefully more accurate forecast before making the decision. One would imagine that the answer to this question should depend on the extent to which there is a benefit in making the decision now rather than later, combined with an understanding of how the skill of the forecast improves, and information about the possible size and nature of forecast changes. We extend the well-known cost-loss model for forecast-based decision making to capture an idealized version of this situation. We find that within this extended cost-loss model, the question of whether to decide now or wait depends on two specific aspects of the forecast, both of which involve probabilities of probabilities. For the special case of weather and climate forecasts in the form of normal distributions we derive a simulation algorithm, and equivalent analytical expressions, for calculating these two probabilities. We apply the algorithm to forecasts of temperature and find that the algorithm leads to better decisions relative to three simpler alternative decision-making schemes. Similar problems have been studied in many other fields, and we explore some of the connections.

The redox properties of quinones underlie their unique characteristics as organic battery components that outperform the conventional inorganic ones. Furthermore, the redox properties could be precisely shaped by using different substituent groups. Machine learning and statistics, onthe other hand, have proven to be very powerful approaches for efficient in silico design of novel materials. Herein, we demonstrated the machine learning approach for the prediction of the redox activity of quinones that potentially can serve as organic battery components. For the needs of the present study, a database of small quinone-derived molecules was created. A large number of quantum chemical and chemometrics descriptors was generated for each molecule and subsequently different statistical approaches were applied to select the descriptors that most prominently characterize the relationship between structure and redox-potential. Various machine-learning methods for screening of prospective organic battery electrode materials were deployed to select the most trustworthy strategy for machine learning aided design of organic redox materials. It was found that ridge regression models perform better than regression decision tree and decision tree based ensemble algorithms.

This paper presents a two-stage hierarchical neural network using image classification and object detection algorithms as key building blocks for a system that automatically detects a potential design right infringement. This neural network is trained to return the Top-N original design right records that highly resemble the input image of a counterfeit. Design rights specify the unique aesthetic characteristics of a product. Due to the rapid change of trends, new design rights are continuously generated. This work proposes an Ensemble Neural Network (ENN), an artificial neural network model that aims to deal with a large amount of counterfeit data and design right records that are frequently added and deleted. At first, we performed image classification and objection detection learning per design right using the existing models with a proven track record of high accuracy. The distributed models form the backbone of the ENN and yield intermediate results aggregated at a master neural network. This master neural network is a deep residual network paired with a fully connected network. This ensemble layer is trained to determine the sub-models that return the best result for a given input image of a product. In the final stage, the ENN model multiples the inferred similarity coefficients to the weighted input vectors produced by the individual sub-models to assess the similarity between the test input image and the existing product design rights to see any sign of violation. Given 84 design rights and the sample product images taken meticulously under various conditions, our ENN model achieved average Top-1 and Top-3 accuracies of 98.409% and 99.460%, respectively. Upon introducing new design rights data, a partial update of the inference model was done an order of magnitude faster than the single model. ENN maintained a high level of accuracy as it scaled out to handle more design rights. Therefore, the ENN model is expected to offer practical help to the inspectors in the field, such as the customs at the border that deal with a swarm of products.

The potential impacts of climate change on the distribution of tuna in Pacific Island Countries’ Exclusive Economic Zones have yet to be investigated rigorously, and so their persistence and abundance in these areas remain uncertain. Here, we estimate optimal fisheries areas for four tuna species; Albacore (Thunnus alalunga), Bigeye (Thunnus obesus), Skipjack (Katsuwonus pelamis), and Yellowfin (Thunnus albacares). We consider different climate change scenarios, RCP 2.6, RCP 4.5, RCP 6.0 and RCP 8.5, within a set of tuna catch records in the Exclusive Economic Zone of Tonga. Using environmental and CPUE datasets, species distribution modelling estimated and predicted these fisheries areas in the current and future climatic scenarios. Our projections indicate an expansion in area and a shift of productive areas to the southern part of this Exclusive Economic Zone of Tonga. This is an indication that future climatic scenarios might be suitable for the species under study however, changes in trophic layers, ocean currents and ocean chemistry might alter this finding. Information provided here will be relevant in planning future national actions towards proper management of these species.

The current study aims to apply and compare the performance of six machine learning algorithms, including three basic classifiers: random forest (RF), gradient boosting decision tree (GBDT), and extreme gradient boosting (XGB), as well as their hybrid classifiers, using the logistic regression (LR) method (RF+LR, GBDT+LR, and XGB+LR), in order to map the landslide susceptibility of Zhangjiajie City, Hunan Province, China. First, a landslide inventory map was created with 206 historical landslide points and 412 non-landslide points, which was randomly divided into two datasets for model training (80%) and model testing (20%). Second, 15 landslide conditioning factors (i.e., altitude, slope, aspect, plane curvature, profile curvature, relief, roughness, rainfall, topographic wetness index (TWI), normalized difference vegetative index (NDVI), distance to roads, distance to rivers, land use/land cover (LULC), soil texture, and lithology) were initially selected to establish a landslide factor database. Thereafter, the multicollinearity test and information gain ratio (IGR) technique were applied to rank the importance of the factors. Subsequently, we used a series of metrics (e.g., accuracy, precision, recall, f-measure, area under the ROC (receiver operating characteristic) curve (AUC), kappa index, mean absolute error (MAE), and root mean square error (RMSE)) to evaluate the accuracy and performance of the six models. Based on the AUC values derived from the models, the GBDT+LR model with the highest AUC value (0.8168) was identified as the most efficient model for mapping landslide susceptibility, followed by the XGB+LR, XGB, RF+LR, GBDT, and RF models, which achieved AUC values of 0.8124, 0.8118, 0.8060, 0.7927, and 0.7883, respectively. The results from this study suggest that the stacking ensemble machine learning method is promising for use in landslide susceptibility mapping in the Zhangjiajie area and is capable of targeting the areas prone to landslides.

Photovoltaic (PV) generation is potentially uncertain. Probabilistic PV generation forecasting methods have been proposed with prediction intervals (PIs). However, several studies have dealt with geographically distributed PVs in a certain area. In this study, a two-step probabilistic forecast scheme is proposed for geographically distributed PV generation forecasting. Each step of the proposed scheme adopts ensemble forecasting based on three different machine-learning methods. In this case study, the proposed scheme was compared with conventional non-multistep forecasting. The proposed scheme improved the reliability of the PIs and deterministic PV forecasting results through 30 days of continuous operation with real data in Japan.

Freezing of Gait (FOG) is an impairment that affects the majority of patients in the advanced stages of Parkinson’s Disease (PD). FOG can lead to sudden falls and injuries, negatively impacting the quality of life for the patients and their families. Rhythmic Auditory Stimulation (RAS) can be used to help patients recover from FOG and resume normal gait. RAS might be ineffective due to the latency between the start of a FOG event, it’s detection and initialization of RAS. We propose a system capable of both FOG prediction and detection using signals from tri-axial accelerometer sensors that will be useful in initializing RAS with minimal latency. We compared the performance of several time frequency analysis techniques, including moving windows extracted from the signals, handcrafted features, Recurrence Plots (RP), Short Time Fourier Transform (STFT), Discreet Wavelet Transform (DWT) and Pseudo Wigner Ville Distribution (PWVD) with Deep Learning (DL) based Long Short Term Memory (LSTM) and Convolutional Neural Networks (CNN). We also propose three Ensemble Network Architectures that combine all the time frequency representations and DL architectures. Experimental results show that our ensemble architectures significantly improve the performance compared with existing techniques. We also present the results of applying our method trained on publicly available dataset to data collected from patients using wearable sensors in collaboration with A.T. Still University.

The efficient recognition and classification of personal protective equipment are essential for en-suring the safety of personnel in complex industrial settings. Using the existing methods, manually performing macro-level classification and identification of personnel in intricate spheres is tedious, time-consuming, and inefficient. The availability of several artificial intelligence models in recent times presents a new paradigm shift in object classification and tracking in complex settings. In this study, several compact and efficient deep learning model architectures are explored, and a new efficient model is constructed by fusing the learning capabilities of the individual, efficient models for better object feature learning and optimal inferencing. The new model construct follows the contributory learning theory whereby each fussed model brings its learned features, which are then combined to obtain a more accurate and rapid model using normalized quantization-aware learning. During the investigation, a separable convolutional driven model was constructed as a base model, and then the various efficient architectures combined for the rapid identification and classification of the various hardhat classes used in complex industrial settings. A remarkable rapid classification and accuracy were recorded with the new resultant model.

Wearable assistant devices play an important role in daily life for people with disabilities. Those who are hearing impaired may face dangers while walking or driving on the road. The major danger is their inability to hear warning sounds from cars or ambulances. Thus, the goal of this study is to develop a wearable assistant device for the hearing impaired to recognize emergency vehicle sirens on the road using edge computing. An EfficientNet-based fuzzy rank-based ensemble model was proposed to classify seven audio sounds, including human vocalizations and emergency vehicle sirens. This model was embedded in an Arduino Nano 33 BLE Sense development board. The audio files were respectively obtained from the CREMA-D dataset and Large Scale Audio dataset of emergency vehicle sirens on the road, with a total number of 8756 files. The seven audio sounds included neutral vocalization, anger vocalization, fear vocalization, happy vocalization, car horn sound, siren sound, and ambulance siren sound. The audio signal was converted into a spectrogram by the short-time Fourier transform as the feature. When one of the car horns, sirens, or ambulance sirens was detected, the wearable assistant device presented alarms through vibration and messages on the OLED panel. The performances of the EfficientNet-based fuzzy rank-based ensemble model in offline computing achieved an accuracy of 97.1%, precision of 97.79%, sensitivity of 96.8%, and specificity of 97.04%. In edge computing, the results were an accuracy of 95.2%, precision of 93.2%, sensitivity of 95.3%, and specificity of 95.1%. Thus, the proposed wearable assistant device has the potential benefit of helping the hearing impaired avoid traffic accidents.

A concept of ensemble averaged stellar reactors is developed to study the dynamics of processes occurring in stars, allocated in the ~ 200 pc solar neighborhood. According to the effective temperature value, four stellar classes are identified, for which the correlation coefficients and standard deviation are counted. The theory of the buoyancy terrestrial elements is generalized to stellar systems. It was suggested that stars are over-heated due to the shift parameters of the nuclear processes occurring inside the stars, which leads to the synthesis of transuranium elements until the achievement of a critical nuclear mass and star explosion. The heavy transuranium elements sink downward and are concentrated in the stellar depth layers. The physical explanation of the existence of the critical Chandrasekhar star limit has been offered. Based on the spatial analysis of overheated stars, it was suggested that the withdrawal of the stellar reactor from the equilibrium state is a consequence of extragalactic compression inside the galaxy arm due to the arm spirality (not to be confused with the spirality of the galaxy itself).

Computers have evolved over the years and as the evolution continues, we have been ushered into an era where high-speed internet has made it possible for devices in our homes, hospital, energy and industry to communicate with each other. This era is what is known as the Internet of Things (IoT). IoT has several benefits in the health, energy, transportation and agriculture sectors of a country’s economy. These enormous benefits coupled with the computational constraint of IoT devices which makes it difficult to deploy enhanced security protocols on them make IoT devices a target of cyber-attacks. One approach that has been used in traditional computing over the years to fight cyber-attacks is Intrusion Detection System (IDS). However, it is practically impossible to deploy IDS meant for traditional computers in IoT environments because of the computational constraint of these devices. In this regard, this study proposes a lightweight IDS for IoT devices using an incremental ensemble learning technique. We used Gaussian Naive Bayes and Hoeffding tree to build our incremental ensemble model. The model was then evaluated on the TON IoT dataset. Our proposed model was compared with other state-of-the-art methods proposed and evaluated using the same dataset. The experimental results show that the proposed model achieved an average accuracy of 99.98\%. We also evaluated the memory consumption of our model which showed that our model achieved a lightweight model status of 650.11KB as the highest memory consumption and 122.38KB as the lowest memory consumption.

Ensembles of Species Distribution Models (SDMs) represent the geographic ranges of pathogen vectors by combining alternative analytical approaches and merging information on vector occurrences with more extensive environmental data. Biased collection data impact SDMs, regardless of the target species but no studies have compared the differences in the distributions predicted by the ensemble models when different sampling frameworks are used for the same species. We compared Ensemble SDMs for two important Ixodid tick vectors, Amblyomma americanum and Ixodes scapularis in mainland Florida, USA, when inputs were either convenience samples of ticks, or collections obtained using the standard protocols promulgated by the U.S. Centers for Disease Control and Prevention. The Ensemble SDMs for the convenience samples and standard surveys showed only a slight agreement (Kappa = 0.060, A. americanum; 0.053, I. scapularis). Convenience sample SDMs indicated A. americanum and I. scapularis should be absent from 34.5% and 30.9% of the state where standard surveys predicted the highest likelihood of occurrence of the respective vectors. Ensemble models from standard surveys predicted 81.4% and 72.5% (A. americanum and I. scapularis) of convenience sample sites. Omission errors by standard survey SDMs, of the convenience collections, frequently were associated with adjacency to at least one SDM or errors in geocoding algorithms that failed to correctly locate convenience samples. These geocoding errors emphasize commonly overlooked needs to explicitly evaluate and improve data quality for vector survey data used in spatial models.

In this work, we investigate the ability of a data assimilation technique and space-borne observations to quantify and monitor changes in nitrogen oxides (NOx) emissions over North-Western Greece for the summers of 2018 and 2019. In this region, four lignite-burning power plants are located. The data assimilation technique, based on the Ensemble Kalman Filter method, is employed to combine space-borne atmospheric observations from the high spatial resolution Sentinel-5 Precursor (S5P) Tropospheric Monitoring Instrument (TROPOMI) and simulations using the LOTOS-EUROS Chemical Transport model. The Copernicus Atmosphere Monitoring Service-Regional European emissions (CAMS-REG, version 4.2) inventory based on year 2015 is used as the a priori in the simulations. Surface measurements of nitrogen dioxide (NO2) from air quality stations operating in the region are compared with the model surface NO2 output using either the a priori (base run) or the a posteriori (assimilated run) NOx emissions. The high biases found between the in situ NO2 measurements and the base run surface NO2 decrease in the assimilated run in most cases. The bias in the station near the largest power plant decreases to 2.0 μg/m3 (2.83 μg/m3) from 10.5 μg/m3 (8.46 μg/m3) in 2019 (2018 respectively). Concerning the estimated annual a posteriori NOx emissions it was found that, for the pixels hosting the two largest power plants, the assimilated run results in emissions decreased by ~40-50% for 2018 compared to 2015, whereas a larger decrease, of ~70% for both power plants, was found for 2019, after assimilating the space-born observations. For the same power plants, the European Pollutant Release and Transfer Register (E-PRTR) reports decreased emissions in 2018 and 2019 compared to 2015 (-35% and -38% in 2018, -62% and -72% in 2019), in good agreement with the estimated emissions. We further compare the a posteriori emissions to the reported energy production of the power plants during the summer of 2018 and 2019. Mean decreases of about -35% and-63% in NOx emissions are estimated for the two larger power plants in summer of 2018 and 2019, respectively, which are supported by similar decreases in the reported energy production of the power plants (~-30% and -70%, respectively).

Local binary pattern is one of the visual descriptors and can be used as a powerful feature extractor for texture classification. In this paper, a novel representation for face recognition is proposed, called it Bilateral Line Local Binary Patterns (BL-LBP). This scheme is an extension of Line Local Binary Patterns descriptors in the statistical learning subspace. The present bilateral descriptors are fused with an ensemble learning of calibrated SVM models. The performance of this scheme is evaluated using 5 standard face databases. It is found that it is robust against illumination variation, diverse facial expressions and head pose variations and its recognition accuracy reaches 98 percent, running on a mobile device with a processing speed of 63 ms per face. Results suggest that our proposed method can be very useful for the vision systems that have limited resources where the computational cost is critical.

Molecular dynamics simulation is currently the theoretical technique eligible to simulate a wide range of systems from soft condensed matter to biological systems. However, of the excellent results that the technique has arrogated, this approach remains computationally expensive, but with the emergence of the new supercomputing technologies bases on graphics processing units graphical processing units-based systems GPUs, the perspective has changed. The GPUs allow performing large and complex simulations at a significantly reduced time. In this work, we present recent innovations in the acceleration of molecular dynamics in GPUs to simulate non-Hamiltonian systems. In particular, we show the performance of measure-preserving geometric integrator in the canonical ensemble, that is, at constant temperature. We provide a validation and performance evaluation of the code by calculating the thermodynamic properties of a Lennard-Jones fluid. Our results are in excellent agreement with reported data reported from literature, which were calculated with CPUs. The scope and limitations for performing simulations of high-throughput MD under rigorous statistical thermodynamics in the canonical ensemble are discussed and analyzed.

Software fault prediction is the very consequent research topic for software quality assurance. Data driven approaches provide robust mechanisms to deal with software fault prediction. However, the prediction performance of the model highly depends on the quality of dataset. Many software datasets suffers from the problem of class imbalance. In this regard, under-sampling is a popular data pre-processing method in dealing with class imbalance problem, Easy Ensemble (EE) present a robust approach to achieve a high classification rate and address the biasness towards majority class samples. However, imbalance class is not the only issue that harms performance of classifiers. Some noisy examples and irrelevant features may additionally reduce the rate of predictive accuracy of the classifier. In this paper, we proposed two-stage data pre-processing which incorporates feature selection and a new Rough set Easy Ensemble scheme. In feature selection stage, we eliminate the irrelevant features by feature ranking algorithm. In the second stage of a new Rough set Easy Ensemble by incorporating Rough K nearest neighbor rule filter (RK) afore executing Easy Ensemble (EE), named RKEE for short. RK can remove noisy examples from both minority and majority class. Experimental evaluation on real-world software projects, such as NASA and Eclipse dataset, is performed in order to demonstrate the effectiveness of our proposed approach. Furthermore, this paper comprehensively investigates the influencing factor in our approach. Such as, the impact of Rough set theory on noise-filter, the relationship between model performance and imbalance ratio etc. comprehensive experiments indicate that the proposed approach shows outstanding performance with significance in terms of area-under-the-curve (AUC).

Heart disease is a global health concern of paramount importance, causing a significant number of fatalities and disabilities. Precise and timely diagnosis of heart disease is pivotal in pre-venting adverse outcomes and improving patient well-being, thereby creating a growing demand for intelligent approaches to predict heart disease effectively. This paper introduces an Ensemble Heuristic-Metaheuristic Feature Fusion Learning (EHMFFL) algorithm for heart disease diagnosis. Within the EHMFFL algorithm, a diverse ensemble learning model is crafted, featuring different feature subsets for each heterogeneous base learner, including support vector machine, K-nearest neighbors, logistic regression, random forest, naive bayes, decision tree, and XGBoost. The primary objective is to identify the most pertinent features for each base learner, leveraging a combined heuristic-metaheuristic approach that integrates the heuristic knowledge of Pearson correlation coefficient with the metaheuristic-driven grey wolf optimizer. The second objective is to aggregate the decision outcomes of the various base learners through ensemble learning, aimed at constructing a robust prediction model. The performance of the EHMFFL algorithm is rigorously assessed using the Cleveland and Statlog datasets yielding remarkable results with an accuracy of 91.8% and 88.9%, respectively, surpassing state-of-the-art machine learning, ensemble learning, and feature selection techniques in heart disease diagnosis. These findings underscore the potential of the EHMFFL algorithm in enhancing diagnostic accuracy for heart disease and providing valuable support to clinicians in making more informed decisions regarding patient care.

A morphological feature estimation method is presented, that can be used to monitor fish from low contrast underwater images and videos. The fish orientation is classified in order to apply a shape alignment technique that is based on the Ensemble of Regression Trees machine learning method. Shape alignment allows the estimation of fish dimensions (length, height, mass) and the location of fish body parts of particular interest such as the eyes and gills. The developed system can be exploited both for fish monitoring in open sea as well as aquacultures where non-invasive, low cost morphological feature extraction is essential. Fish health can be assessed from malformations in body shape, the color of the eyes or the gills as well as the fish behavior. The proposed method can estimate the position of 18 landmarks with an accuracy of about 95% from low resolution and low contrast underwater images where the fish can be hardly distinguished from its background. Hardware acceleration can be applied at various stages of the proposed method for real time video processing. As a case study, the developed system has been trained and tested with several Mediterranean fish species in the category of seabream or similar: diplodous sargus annularis (white seabream), diplodus annularis seabream (spawn), oblada melanura (saddled seabream), pagrus pagrus (common seabream), etc. A large dataset with low resolution underwater videos and images has been created in the context of this work.

This paper presents ensemble learning of multi-source satellite sensors dataset to obtain better predictive performance of the forest biomass. Spectral, spectral-indices, and spectral-textural features were generated from two optical satellite sensors, Landsat 8 Operational Land Imager (OLI) and Sentinel-2 Multispectral Instrument (MSI). In addition, two radar satellite sensors, Sentinel-1 C-band Synthetic Aperture Radar (CSAR), and Advanced Land Observing Satellite (ALOS-2) Phased Array type L-band Synthetic Aperture Radar (PALSAR-2) were utilized to generate backscattering and backscattering-textural features. The plot-wise above ground biomass data available from five forests in New England region were utilized. Ensemble learning of multi-source satellite sensors dataset was carried out by employing four machine learning regressors namely, Support Vector Machines (SVM), Random Forests (RF), Gradient Boosting (GB), and Multilayer Perceptron (MLP). A five-fold cross-validation method was used to evaluate predictive performance of the multi-source satellite sensors. The integration of multi-source satellite features, comprising of spectral, spectral-indices, backscattering, spectral-textural, and backscattering-textural information, through ensemble learning and cross-validation approach implemented in the research showed promising results (R² = 0.81, RMSE = 46.2 Mg/ha) for the estimation of plots-level forest biomass in New England region.

The progressive impairment analysis in gait from neurological diseases patients such as Hereditary Ataxias (HA) has been carried out using gait data collected with movement sensors. This research is focused on finding the minimum amount required of gait features to recognize efficiently and less intrusive way, HA patients based on data collected with iPhone movement sensors placed on the ankles from 14 HA patients and 14 healthy people. A twofold proposal is made , first a local minimum prominent peak criterion to find out the starting point of each stride, to get 10-stride window about which 56 spatial-temporal features are derived; second a search strategy based on Hill Climbing algorithm to reduce the number of gait features and sensors. The main results were the findings that with two gait patterns a 96% of classification accuracy was achieved by using K-Nearest Neighbors (KNN) and Multi-Layer Perceptron (MLP) algorithms, but in addition, MLP only right ankle sensor patterns were required which also allows to reduce the intrusion.

Automated medical diagnosis is one of the important machine learning applications in the domain of healthcare. In this regard, most of the approaches primarily focus on optimizing the accuracy of classification models. In this research, we argue that unlike general-purpose classification problems, medical applications, such as chronic kidney disease (CKD) diagnosis, require special treatment. In the case of CKD, apart from model performance, other factors such as the cost of data acquisition may also be taken into account to enhance the applicability of the automated diagnosis system. In this research, we have proposed two techniques for cost-sensitive feature ranking. An ensemble of decision tree models is employed in both the techniques for computing the worth of a feature in the CKD dataset. An automatic threshold selection heuristic is also introduced which is based on the intersection of features’ worth and their accumulated cost. A set of experiments are conducted to evaluate the efficacy of the proposed techniques on both tree-based and non-tree based classification models. The proposed approaches are also evaluated against several comparative techniques. Furthermore, it is demonstrated that the proposed techniques select around 1/4^th of the original CKD features while reducing the cost by a factor of 7.42 of the original feature set. Based on the extensive experimentation it is concluded that the proposed techniques employing feature-cost interaction heuristic tend to select feature subsets that are both useful and cost-effective.

To understand functions of biomolecules such as proteins, not only structures but their conformational change and kinetics are important to be characterized but its atomistic details are hard to obtain both experimentally and computationally. We review our recent computational studies using novel enhanced sampling techniques for conformational sampling of biomolecules and calculations of their kinetics. For efficiently characterizing the free energy landscape of a biomolecule, we introduce the multiscale enhanced sampling method, which uses a combined system of atomistic and coarse-grained models. Based on the idea of Hamiltonian replica exchange, we can recover the statistical properties of the atomistic model without any biases. We next introduce the string method as a path search method to calculate the minimum free energy pathways along a multidimensional curve in high dimensional space. Finally we introduce novel methods to calculate kinetics of biomolecules based on the ideas of path sampling: One is the Onsager-Machlup action method, and the other is the weighted ensemble method. Some applications of above methods to biomolecular systems are also discussed and illustrated.

Multilabel learning goes beyond standard supervised learning models by associating a sample with more than one class label. Among the many techniques developed in the last decade to handle multilabel learning best approaches are those harnessing the power of ensembles and deep learners. This work proposes merging both methods by combining a set of gated recurrent units, temporal convolutional neural networks, and long short-term memory networks trained with variants of the Adam optimization approach. We examine many Adam variants, each fundamentally based on the difference between present and past gradients, with step size adjusted for each parameter. We also combine Incorporating Multiple Clustering Centers and a bootstrap-aggregated decision trees ensemble, which is shown to further boost classification performance. In addition, we provide an ablation study for assessing the performance improvement that each module of our ensemble produces. Multiple experiments on a large set of datasets representing a wide variety of multilabel tasks demonstrate the robustness of our best ensemble, which is shown to outperform the state-of-the-art. The MATLAB code for generating the best ensembles in the experimental section will be made available at https://github.com/LorisNanni.

Soil temperature is a fundamental parameter in water resources and engineering. A cost-effective model which can forecast soil temperature accurately is extensively needed. Recently, many studies have applied artificial intelligence (AI) at both surface and underground levels for soil temperature prediction. However, there is no comprehensive and detailed assessment of the performance of different AI approaches in soil temperature estimation, and primarily limited atmospheric variables are used as input data for AI models. In the present study, great varieties of various land and atmospheric variables are applied to evaluate the performance of a wide range of AI methods on soil temperature prediction. Herein, thirteen approaches, from classic regressions to well-established methods of random forest and gradient boosting to advanced AI techniques like multi-layer perceptron and deep learning are taken into account. The results show that AI is a promising approach in climate parameter forecast and deep learning demonstrates the best performance among other models. It has the highest R-squared ranging from 0.957 to 0.980, the lowest NRMSE ranging from 2.237% to 3.287% and the lowest MAE, ranging from 0.510 to 0.743 in predicting soil temperature. The prediction is repeated for different sizes of data, and prediction outcomes confirm the conclusion mentioned above.

Wearable devices have become ubiquitous, collecting rich temporal data that offers valuable insights into human activities, health monitoring, and behavior analysis. Leveraging this data, researchers have developed innovative approaches to classify and predict time-based patterns and events in human life. Time-based techniques allow the capture of intricate temporal dependencies, which is the nature of the data coming from wearable devices. This paper focuses on predicting well-being factors, such as stress, anxiety, positive and negative affect, on the Tesserae dataset collected from office workers. We examine the performance of different methodologies, including deep learning architectures, LSTM, ensemble techniques, Random Forest (RF) and XGBoost and compare their performances for time-based and non-time-based versions. In time-based versions, we investigate the effect of previous records of well-being factors on the upcoming ones. The overall results show that time-based LSTM performs the best among conventional (non-time-based) RF, XGBoost, and LSTM. The performance even increases when we consider a more extended previous period, in this case, 3 past-days rather than 1 past-day to predict the next day. Furthermore, we explore the corresponding biomarkers for each well-being factor using feature ranking. The obtained rankings are compatible with the psychological literature. In this work, we validated them based on device measurements rather than subjective survey responses.

A machine learning method for classifying Lung UltraSound is here proposed to pro- vide a point of care tool for supporting a safe, fast and accurate diagnosis, that can also be useful during a pandemic like as SARS-CoV-2. Given the advantages (e.g. safety, rapidity, portability, cost-effectiveness) provided by the ultrasound technology over other methods (e.g. X-ray, computer tomography, magnetic resonance imaging), our method was validated on the largest LUS public dataset. Focusing on both accuracy and efficiency, our solution is based on an efficient adaptive ensembling of two EfficientNet-b0 models reaching 100% of accuracy, which, to our knowledge, outperforms the previous state-of-the-art. The complexity of this solution keeps the number of parameters in the same order as an EfficientNet-b0 by adopting specific design choices that are adaptive ensembling with a combination layer, ensembling performed on the deep features, minimal ensemble only two weak models. Moreover, a visual analysis of the saliency maps on sample images of all the classes of the dataset reveals where the focus is on an inaccurate weak model versus an accurate model.

Automated lying-posture tracking is important in preventing bed-related disorders such as pressure injuries, sleep apnea, and lower-back pain. Prior research studied in-bed lying posture tracking using sensors of different modalities (e.g., accelerometer and pressure sensors). However, there remain significant gaps in research about how to design efficient in-bed lying posture tracking systems. These gaps can be articulated through several research questions as follows. First, can we design a single-sensor, pervasive, and inexpensive system that can accurately detect lying postures? Second, what computational models are most effective in the accurate detection of lying postures? Finally, what physical configuration of the sensor system is most effective for lying posture tracking? To answer these important research questions, in this article, we propose a comprehensive approach to design a sensor system that uses a single accelerometer along with machine learning algorithms for in-bed lying posture classification. We design two categories of machine learning algorithms based on deep learning and traditional classification with handcrafted features to detect lying postures. We also investigate what wearing sites are most effective in accurate detection of lying postures. We extensively evaluate the performance of the proposed algorithms on nine different body locations and four human lying postures using two datasets. Our results show that a system with a single accelerometer can be used with either deep learning or traditional classifiers to accurately detect lying postures. The best models in our approach achieve an F-Score that ranges from 95.2% to 97.8% with a coefficient of variation from 0.03 to 0.05. The results also identify the thighs and chest as the most salient body sites for lying posture tracking. Our findings in this article suggest that because accelerometers are ubiquitous and inexpensive sensors, they can be a viable source of information for pervasive monitoring of in-bed postures.

Antisense oligonucleotide (ASO)-mediated exon skipping has become a valuable tool for investigating gene function and developing gene therapy. Machine learning-based computational methods such as eSkip-Finder have been developed to predict the efficacy of ASOs via exon skipping. However, these methods are computationally demanding, and the accuracy of predictions remains suboptimal. In this study, we propose a new approach to reduce computational burden and improve prediction performance by using feature selection within machine learning algorithms and ensemble learning techniques. We evaluated our approach using a dataset of experimentally validated exon skipping events, dividing it into training and testing sets. Our results demonstrate that using a 3-way voting approach with random forest, gradient boosting, and XGBoost can significantly reduce computation time to under ten seconds while improving prediction performance, as measured by R2 for both 2’-O-methyl nucleotides (2OMe) and phosphorodiamidate morpholino oligomers (PMOs). Additionally, the feature importance ranking derived from our approach is in good agreement with previously published results. Our findings suggest that our approach has the potential to enhance the accuracy and efficiency of predicting ASO efficacy via exon skipping. It could also facilitate the development of novel therapeutic strategies. This study could contribute to the ongoing efforts to improve ASO design and optimize gene therapy approaches.

Antisense oligonucleotide (ASO)-mediated exon skipping has emerged as a powerful tool for examining the function of genes and exons in basic research, as well as gene therapy. Computational methods, such as eSkip-Finder, have been developed to predict the efficacy of ASOs via exon skipping using machine learning. However, these methods can be computationally demanding and the prediction accuracy of the tool is not yet optimal. In this study, we propose an approach to reduce computational burden and improve prediction performance by utilizing feature selection within machine learning algorithms and employing ensemble learning techniques. The method was evaluated using a dataset of genes with experimentally validated exon skipping events. The dataset was divided into training and testing sets to assess the accuracy of the algorithm. Our results demonstrate that using a 3-way voting approach with random forest, gradient boosting, and XGBoost can significantly reduce computation time to under ten seconds while improving prediction performance, as measured by R2 for both 2’-O-methyl nucleotides (2OMe) and phosphorodiamidate morpholino oligomers (PMOs). Additionally, the feature importance ranking derived from our approach is in good agreement with previously published results. These findings suggest that this approach has the potential to enhance the efficiency and accuracy of predicting ASO efficacy via exon skipping, facilitating the development of novel therapeutic strategies.