Ensemble Model Research Articles

An Efficient Ensemble Approach for Brain Tumors Classification Using Magnetic Resonance Imaging

Tumors in the brain can be life-threatening, making early and precise detection crucial for effective treatment and improved patient outcomes. Deep learning (DL) techniques have shown significant potential in automating the early diagnosis of brain tumors by analyzing magnetic resonance imaging (MRI), offering a more efficient and accurate approach to classification. Deep convolutional neural networks (DCNNs), which are a sub-field of DL, have the potential to analyze rapidly and accurately MRI data and, as such, assist human radiologists, facilitating quicker diagnoses and earlier treatment initiation. This study presents an ensemble of three high-performing DCNN models, i.e., DenseNet169, EfficientNetB0, and ResNet50, for accurate classification of brain tumors and non-tumor MRI samples. Our proposed ensemble model demonstrates significant improvements over various evaluation parameters compared to individual state-of-the-art (SOTA) DCNN models. We implemented ten SOTA DCNN models, i.e., EfficientNetB0, ResNet50, DenseNet169, DenseNet121, SqueezeNet, ResNet34, ResNet18, VGG16, VGG19, and LeNet5, and provided a detailed performance comparison. We evaluated these models using two learning rates (LRs) of 0.001 and 0.0001 and two batch sizes (BSs) of 64 and 128 and identified the optimal hyperparameters for each model. Our findings indicate that the ensemble approach outperforms individual models, having 92% accuracy, 90% precision, 92% recall, and an F1 score of 91% at a 64 BS and 0.0001 LR. This study not only highlights the superior performance of the ensemble technique but also offers a comprehensive comparison with the latest research.

An efficient interpretable stacking ensemble model for lung cancer prognosis

Lung cancer significantly contributes to global cancer mortality, posing challenges in clinical management. Early detection and accurate prognosis are crucial for improving patient outcomes. This study develops an interpretable stacking ensemble model (SEM) for lung cancer prognosis prediction and identifies key risk factors. Using a Kaggle dataset of 1000 patients with 22 variables, the model classifies prognosis into Low, Medium, and High-risk categories. The bootstrap method was employed for evaluation metrics, while SHAP (Shapley Additive Explanations) and LIME (Local Interpretable Model-agnostic Explanations) assessed model interpretability. Results showed SEM's superior interpretability over traditional models, such as Random Forest, Logistic Regression, Decision Tree, Gradient Boosting Machine, Extreme Gradient Boosting Machine, and Light Gradient Boosting Machine. SEM achieved an accuracy of 98.90 %, precision of 98.70 %, F1 score of 98.85 %, sensitivity of 98.77 %, specificity of 95.45 %, Cohen’s kappa value of 94.56 %, and an AUC of 98.10 %. The SEM demonstrated robust performance in lung cancer prognosis, revealing chronic lung cancer and genetic risk as major factors.

Feasibility study of data-driven wall interference correction framework for subsonic wind tunnel

Although the classical method is widely used for wall interference correction in wind tunnel testing, its reliability and accuracy for complex and unconventional geometries are rather limited. Studies on the evaluation of wall interference and the improvement of correction methods are desirable to enhance the reliability and generality for various geometric configurations. This study proposes a wall interference correction framework based on a deep neural network (DNN) ensemble using data obtained from the numerical panel method. The panel method is validated by comparing the results with those of Reynolds-averaged Navier-Stokes simulations. An automated process was established to generate a large amount of training data, and 600,000 datasets were generated based on the geometric parameters of the wind tunnel, test model, and angles of attack. The input variables of the DNN were determined through sensitivity analysis of the data. To alleviate the randomness of the initial weights and data distribution in the generation process of the DNN model, 20 DNNs with the same multi-layer perceptron structure were trained, and a DNN ensemble model was constructed using five ensemble members with high predictability. The accuracy of the DNN-ensemble based correction models were evaluated by comparing the correction results for the testing data.

LLM4THP: a computing tool to identify tumor homing peptides by molecular and sequence representation of large language model based on two-layer ensemble model strategy

Tumor homing peptides (THPs) have a distinctive capacity to specifically attach to tumor cells, providing a promising approach for targeted cancer treatment and detection. Although THPs have the potential for significant impact, their detection by conventional methods is both time-consuming and expensive. To tackle this issue, we provide LLM4THP, an innovative computational approach that utilizes large language models (LLMs) to quickly and effectively detect THPs. LLM4THP utilizes two protein LLMs, ESM2 and Prot_T5_XL_UniRef50, to encode peptide sequences. This allows for the capture of complex patterns and relationships within the peptide data. In addition, we utilize inherent sequence characteristics such as Amino Acid Composition (AAC), Pseudo Amino Acid Composition (PAAC), Amphiphilic Pseudo Amino Acid Composition (APAAC), and Composition, Transition, and Distribution (CTD) to improve the representation of peptides. The RDKitDescriptors feature representation approach transforms peptide sequences into molecular objects and computes chemical characteristics, resulting in enhanced THP identification. The LLM4THP ensemble strategy incorporates various features into a two-layer learning architecture. The first layer consists of LightGBM, XGBoost, Random Forest, and Extremely Randomized Trees, which generate a set of meta results. The second layer utilizes Logistic Regression to further refine the identification of sequences as either THP or non-THP. LLM4THP exhibits exceptional performance compared to the most advanced methods, showcasing enhancements in accuracy, Matthew’s correlation coefficient, F1 score, area under the curve, and average precision. The source code and dataset can be accessed at the following URL: https://github.com/abcair/LLM4THP.

Multi-modal classification of breast cancer lesions in Digital Mammography and contrast enhanced spectral mammography images

Breast cancer ranks as the second most prevalent cancer in women, recognized as one of the most dangerous types of cancer, and is on the rise globally. Regular screenings are essential for early-stage treatment. Digital mammography (DM) is the most recognized and widely used technique for breast cancer screening. Contrast-Enhanced Spectral Mammography (CESM or CM) is used in conjunction with DM to detect and identify hidden abnormalities, particularly in dense breast tissue where DM alone might not be as effective. In this work, we explore the effectiveness of each modality (CM, DM, or both) in detecting breast cancer lesions using deep learning methods. We introduce an architecture for detecting and classifying breast cancer lesions in DM and CM images in Craniocaudal (CC) and Mediolateral Oblique (MLO) views. The proposed architecture (JointNet) consists of a convolution module for extracting local features, a transformer module for extracting long-range features, and a feature fusion layer to fuse the local features, global features, and global features weighted based on the local ones. This significantly enhances the accuracy of classifying DM and CM images into normal or abnormal categories and lesion classification into benign or malignant. Using our architecture as a backbone, three lesion classification pipelines are introduced that utilize attention mechanisms focused on lesion shape, texture, and overall breast texture, examining the critical features for effective lesion classification. The results demonstrate that our proposed methods outperform their components in classifying images as normal or abnormal and mitigate the limitations of independently using the transformer module or the convolution module. An ensemble model is also introduced to explore the effect of each modality and each view to increase our baseline architecture's accuracy. The results demonstrate superior performance compared with other similar works. The best performance on DM images was achieved with the semi-automatic AOL Lesion Classification Pipeline, yielding an accuracy of 98.85 %, AUROC of 0.9965, F1-score of 98.85 %, precision of 98.85 %, and specificity of 98.85 %. For CM images, the highest results were obtained using the automatic AOL Lesion Classification Pipeline, with an accuracy of 97.47 %, AUROC of 0.9771, F1-score of 97.34 %, precision of 94.45 %, and specificity of 97.23 %. The semi-automatic ensemble AOL Classification Pipeline provided the best overall performance when using both DM and CM images, with an accuracy of 94.74 %, F1-score of 97.67 %, specificity of 93.75 %, and sensitivity of 95.45 %. Furthermore, we explore the comparative effectiveness of CM and DM images in deep learning models, indicating that while CM images offer clearer insights to the human eye, our model trained on DM images yields better results using Attention on Lesion (AOL) techniques. The research also suggests a multimodal approach using both DM and CM images and ensemble learning could provide more robust classification outcomes.

Artificial intelligence driven cyberattack detection system using integration of deep belief network with convolution neural network on industrial IoT

In Industry 4.0, information and communication technology (ICT) was employed in numerous significant infrastructures, like financial networks, smart factories, and power plants, to automate and certify industrial systems. In power control systems, ICT technologies such as IIoT have improved automated monitoring, but legacy methods, originally autonomous, now connect with external networks. This progress has presented safety vulnerabilities from legacy ICT systems. Hence, various cybersecurity approaches are developed and examined to deal with cyberattacks and vulnerabilities. Utilizing new cybersecurity models in power control systems poses risks due to their uncertified safety. Ensuring their stability and efficiency is significant for maintaining reliable power delivery and incorporating these technologies into power control systems. Therefore, this study designs a Next–Generation Cybersecurity Attack Detection using an ensemble deep learning model (NGCAD-EDLM) technique in the IIoT environment. The main cause of the NGCAD-EDLM technique is the automatic recognition of cyber-attacks. In the NGCAD-EDLM approach, the primary data normalization phase utilizing min-max normalization is performed. Next, the honey-badger algorithm (HBA) approach selects the feature subsets. Furthermore, an ensemble deep learning (DL) of two methods, namely convolutional neural networks (CNNs) and deep belief networks (DBNs) methods, are employed for classification. In addition, the DL techniques' hyperparameter selection is accomplished using the lotus effect optimization algorithm (LEOA) method. A complete set of simulation validation is performed to establish the experimental analysis of the NGCAD-EDLM method. The performance validation of the NGCAD-EDLM method exhibited a superior accuracy value of 99.21 % over other existing techniques.

Ensemble Learning Development Based on Transfer Learning for Indonesian Traditional Food Detection

Development of traditional food competes with other traditional foods now. They must compete with fast food and food from abroad. In 2013, the food and beverage sector were the second highest contributor to tourist expenditure after accommodation. This shows its very important role in the economy. That caused, we need a model that can predict traditional Indonesian foods and snacks. We used ensemble learning. It had 2 transfer learning methods, namely VGG-19 and Xception. They will be combined to improve the performance of the existing model. The research result shown output. It has found that the ensemble learning model achieved accuracy of up to 97% on training data and 91% on testing data. It is hoped that this prediction model can help people recognize typical Indonesian food and increase interest in and preserve the food around them.

2D Transdimensional Joint Inversion of Radio Magnetotelluric and Electrical Resistivity Tomography Data

Summary The joint inversion of radio magnetotelluric and electrical resistivity tomography data has the potential to reduce the uncertainties in the subsurface conductivity model. This is particularly beneficial when the datasets offer complementary information about the subsurface. However, the traditional gradient-based inversion methods pose challenges in quantifying uncertainty, as they yield a single model with limited appraisal of parameter uncertainty. The Bayesian inversion approach stands out for its capacity to provide quantitative assessments of uncertainty in the inverted model parameters. This is accomplished by generating an ensemble of models, leading to a posterior distribution that encapsulates both prior information concerning model parameters and the dataset information. We have implemented a transdimensional Markov chain Monte Carlo algorithm to perform the joint inversion of radio magnetotelluric and electrical resistivity tomography data. Through synthetic data studies, we illustrate how the inclusion of two complementary datasets can effectively reduce uncertainties in model parameters and how the model parameter uncertainties can be quantified. Subsequently, the developed algorithm is tested using exemplary field data from a waste site near Roorkee, India. Intensive prior geoelectric and transient electromagnetic as well as radio magnetotelluric studies investigated possible waste water seepage with a potential to contaminate the shallow aquifers. The derived subsurface structure from our transdimensional Bayesian results compare well with the deterministic results for the exemplary profile, but in addition provide comprehensive uncertainty estimates.

Ensemble Learning for Stellar Classification and Radius Estimation from Multimodal Data

Abstract Stellar classification and radius estimation are crucial for understanding the structure of the universe and stellar evolution. With the advent of the era of astronomical big data, multimodal data are available and theoretically effective for stellar classification and radius estimation. A problem is how to improve the performance of this task by jointly using the multimodal data. However, existing research primarily focuses on using single-modal data. To this end, this paper proposes a model, Multi-Modal SCNet (MMSCNet), and its ensemble model Multimodal Ensemble for Stellar Classification and Regression (MESCR) for improving stellar classification and radius estimation performance by fusing two modality data. In this problem, a typical phenomenon is that the sample numbers of some type stars are evidently more than others. This imbalance has negative effects on model performance. Therefore, this work utilize a weighted sampling strategy to deal with the imbalance issues in MESCR. Some evaluation experiments are conducted on a test set for MESCR and the classification accuracy is 96.1\%, the radius estimation performance Mean of Absolute Error (MAE) and $\sigma$ are 0.084 dex and 0.149 \(R_{\odot}\) respectively. Moreover, we assessed the uncertainty of model predictions, confirming good consistency within a reasonable deviation range. Finally, we applied our model to 50,871,534 SDSS stars without spectra and published a new catalog.

AddiVortes: (Bayesian) Additive Voronoi Tessellations

The Additive Voronoi Tessellations (AddiVortes) model is a multivariate regression model that uses Voronoi tessellations to partition the covariate space in an additive ensemble model. Unlike other partition methods, such as decision trees, this has the benefit of allowing the boundaries of the partitions to be non-orthogonal and non-parallel to the covariate axes. The AddiVortes model uses a similar sum-of-tessellations approach and a Bayesian backfitting MCMC algorithm to the BART model. We utilize regularization priors to limit the strength of individual tessellations and accepts new models based on a likelihood. The performance of the AddiVortes model is illustrated through testing on several data sets and comparing the performance to other models along with a simulation study to verify some of the properties of the model. In many cases, the AddiVortes model outperforms random forests, BART and other leading black-box regression models when compared using a range of metrics. Supplementary materials for this article are available online.

MisORFPred: A Novel Method to Mine Translatable sORFs in Plant Pri-miRNAs Using Enhanced Scalable k-mer and Dynamic Ensemble Voting Strategy.

The primary microRNAs (pri-miRNAs) have been observed to contain translatable small open reading frames (sORFs) that can encode peptides as an independent element. Relevant studies have proven that those of sORFs are of significance in regulating the expression of biological traits. The existing methods for predicting the coding potential of sORFs frequently overlook this data or categorize them as negative samples, impeding the identification of additional translatable sORFs in pri-miRNAs. In light of this, a novel method named misORFPred has been proposed. Specifically, an enhanced scalable k-mer (ESKmer) that simultaneously integrates the composition information within a sequence and distance information between sequences is designed to extract the nucleotide sequence features. After feature selection, the optimal features and several machine learning classifiers are combined to construct the ensemble model, where a newly devised dynamic ensemble voting strategy (DEVS) is proposed to dynamically adjust the weights of base classifiers and adaptively select the optimal base classifiers for each unlabeled sample. Cross-validation results suggest that ESKmer and DEVS are essential for this classification task and could boost model performance. Independent testing results indicate that misORFPred outperforms the state-of-the-art methods. Furthermore, we execute misORFPerd on the genomes of various plant species and perform a thorough analysis of the predicted outcomes. Taken together, misORFPred is a powerful tool for identifying the translatable sORFs in plant pri-miRNAs and can provide highly trusted candidates for subsequent biological experiments.

PREDICTION INDONESIA COMPOSITE INDEX USING INTEGRATION DECOMPOSITION- NEURAL NETWORK ENSEMBLE DURING VUCA ERA

The Volatility, Uncertainty, Complexity, and Ambiguity (VUCA) era causes turmoil in the capital markets, stocks, commodities, etc. The impact is a decline in the Composite Stock Price Index (IHSG) in 2020. Therefore, future data is needed to inform investors and business people when making portfolio decisions. This paper develops a decomposition and Neural Network (NN) integration model to predict ICI during the VUCA era. The results are presented empirically to show the model's effectiveness in reducing prediction errors. First, the actual data is converted into three components; second, with the Neural Network Ensemble (NNE) approach where the initial step of decomposition results is trained using artificial NN with architecture, training data, and topology to produce individual networks; The output is selected using Principal Component Analysis (PCA) and becomes input to the ensemble model, then combined using a simple average and weighted average. The empirical results from ICI predictions illustrate: (1) decomposition has the potential to overcome data that is characterized by high volatility; (2) NNE is able to reduce errors (MSE≤0.100e-4, MAE≤0.01) compared to individual networks (MSE=0.0024 MAE=0.0376); (3) ensemble combinations using weighted averages (MSE≤3.00e-5,MAE≤0.002) are superior to simple averages (MSE≤5.00e-5,MAE≤0.01); (4) the integration carried out shows effectiveness in predicting ICI and provides better prediction results.

Bolt loosening assessment using ensemble vision models for automatic localization and feature extraction with target‐free perspective adaptation

AbstractBolt loosening assessment is crucial to identify early warnings of structural degradation and prevent catastrophic events. This paper proposes an automatic bolt loosening assessment methodology. First, a novel end‐to‐end ensemble vision model, Bolt‐FP‐Net, is proposed to reason the locations of bolts and their hexagonal feature patterns concurrently. Second, an adaptive target‐free perspective correction method is proposed to correct perspective distortion and enhance assessment accuracy. Finally, an iterative bolt loosening quantification is developed to estimate and refine the bolt loosening rotation. Experimental parametric studies indicated that the proposed Bolt‐FP‐Net can achieve excellent performance under different environmental conditions. Finally, a case study was conducted on steel bolt connections, which shows the proposed methodology can achieve high accuracy and real‐time speed in bolt loosening assessment.

Long-term exposure to PM2.5 and mortality: a national health insurance cohort study.

Previous studies with large data have been widely reported that exposure to fine particulate matter (PM2.5) is associated with all-cause mortality; however, most of these studies adopted ecological time-series designs or have included limited study areas or individuals residing in well-monitored urban areas. However, nationwide cohort studies including cause-specific mortalities with different age groups were sparse. Therefore, this study examined the association between PM2.5 and cause-specific mortality in South Korea using the nationwide cohort. A longitudinal cohort with 187917 National Health Insurance Service-National Sample Cohort participants aged 50-79 years in enrolment between 2002 and 2019 was used. Annual average PM2.5 was collected from a machine learning-based ensemble model (a test R2 = 0.87) as an exposure. We performed a time-varying Cox regression model to examine the association between long-term PM2.5 exposure and mortality. To reduce the potential estimation bias, we adopted generalized propensity score weighting method. The association with long-term PM2.5 (2-year moving average) was prominent in mortalities related to diabetes mellitus [hazard ratio (HR): 1.03 (95% CI: 1.01, 1.06)], circulatory diseases [HR: 1.02 (95% CI: 1.00, 1.03)] and cancer [HR: 1.01 (95% CI: 1.00, 1.02)]. Meanwhile, circulatory-related mortalities were associated with a longer PM2.5 exposure period (1 or 2-year lags), whereas respiratory-related mortalities were associated with current-year PM2.5 exposure. In addition, the association with PM2.5 was more evident in people aged 50-64 years than in people aged 65-79 years, especially in heart failure-related deaths. This study identified the hypothesis that long-term exposure to PM2.5 is associated with mortality, and the association might be different by causes of death. Our result highlights a novel vulnerable population: the middle-aged population with risk factors related to heart failure.

An adversarial diverse deep ensemble approach for surrogate‐based traffic signal optimization

AbstractSurrogate‐based traffic signal optimization (TSO) is a computationally efficient alternative to simulation‐based TSO. By replacing the simulation‐based objective function, a surrogate model can quickly identify solutions by searching for extreme points on its response surface. As a popular surrogate model, the ensemble of multiple diverse deep learning models can approximate complicated systems with a strong generalizability. However, existing ensemble methods barely focus on strengthening the prediction of extreme points, which we found can be realized by further diversifying base learners in the ensemble. The study proposes an adversarial diverse ensemble (ADE) method for online TSO with limited computational resources, comprising two stages: In the offline stage, base extractors are diversified with unlabeled data by a designed adversarial diversity training algorithm; in the online stage, base predictors are trained in parallel with limited labeled data, and the ensemble then serves as the surrogate model to search for solutions iteratively for TSO. First, it is demonstrated that the prediction accuracy on extreme points, and associated solution quality, can be constantly improved with base learners’ diversity enhanced by ADE. Case studies of TSO conducted on a four‐intersection arterial further demonstrate the superior solution quality and computational efficiency of the ADE surrogate model in a wide range of traffic scenarios. Moreover, a large‐scale online TSO experiment under dynamic traffic demand proves ADE's effectiveness in practical applications.

Epidemiology and Ecology of Usutu Virus Infection and Its Global Risk Distribution.

Usutu virus (USUV) is an emerging mosquito-transmitted flavivirus with increasing incidence of human infection and geographic expansion, thus posing a potential threat to public health. In this study, we established a comprehensive spatiotemporal database encompassing USUV infections in vectors, animals, and humans worldwide by an extensive literature search. Based on this database, we characterized the geographic distribution and epidemiological features of USUV infections. By employing boosted regression tree (BRT) models, we projected the distributions of three main vectors (Culex pipiens, Aedes albopictus, and Culiseta longiareolata) and three main hosts (Turdus merula, Passer domesticus, and Ardea cinerea) to obtain the mosquito index and bird index. These indices were further incorporated as predictors into the USUV infection models. Through an ensemble learning model, we achieved a decent model performance, with an area under the curve (AUC) of 0.992. The mosquito index contributed significantly, with relative contributions estimated at 25.51%. Our estimations revealed a potential exposure area for USUV spanning 1.80 million km2 globally with approximately 1.04 billion people at risk. This can guide future surveillance efforts for USUV infections, especially for countries located within high-risk areas and those that have not yet conducted surveillance activities.

An Automated Period Prediction System for Sinhala Epigraphical Scripts using Ensemble CNNs and Attention Modules

Identifying the period of epigraphical scripts is crucial for archaeologists and others to determine the age of inscriptions. Since different sets and shapes of letters were used in different eras, identifying the period of an epigraphical script also aids in recognizing these scripts. An ideal period prediction system should detect the era of an epigraphical script in real time with high accuracy. The objective of this study is to develop an automated system to predict the period of Sri Lankan Sinhala epigraphical scripts. In the first stage, a dataset of Sinhala epigraphical letter images was created using 7,012 samples from estampage pictures of Sri Lankan inscriptions, addressing the absence of a proper dataset. The proposed approach is more efficient than previous models as it can detect the period of individual letters as well as the period of raw, whole estampage images. Moreover, the approach incorporates a mechanism to detect the period of letters from inscriptions written between two consecutive eras. An ensemble CNN model with attention modules is utilized to identify the eras of epigraphical scripts. Experimental results show that the proposed system achieves an average classification accuracy of 93.88% in identifying the era of individual letters. The system can automatically determine the era of an inscription by analyzing its estampage image within thirty seconds.

Engine Mass Flow Estimation through Neural Network Modeling in Semi-Transient Conditions: A New Calibration Approach

Nowadays, engine experimental research represents a very expensive field within the automotive industry, but it remains fundamental for engine and vehicle development. The present work aims to investigate a novel approach for engine control system calibration, by adopting machine learning techniques to model physical parameters of the engine starting from experimental data measured at the test bench. The main goal is to create a methodology which accelerates the calibration process without losing accuracy. A model that estimates air mass flow is created by adopting either a tree ensemble model or an artificial neural network trained on a small dataset, which was previously acquired at the test bench using a random calibration of the volumetric efficiency map. The model’s performance is first validated on a larger, random dataset. Then, the volumetric efficiency calculated from the air mass flow model estimation is used to calibrate the transfer function of the Engine Control Unit. Finally, the sensitivity of the model error correlated with the number of data points acquired is used in order to determine the best practice for a Design Of Experiment, which minimizes data acquisition. The methodology proposed can lead to reduced time and costs of the whole calibration process of the engine, without losing accuracy. The analysis was conducted on the entire vehicle, which is crucial for drivability, especially in motorcycles since they are highly sensitive to air-to-fuel ratio adjustments. This work demonstrates that machine learning models can be adopted for the fine-tuning of the calibration process, which is normally performed manually.

Evaluating ICESat-2 and GEDI with Integrated Landsat-8 and PALSAR-2 for Mapping Tropical Forest Canopy Height

Mapping forest canopy height is critical for climate modeling and forest management, and tropical forests present unique challenges for remote sensing due to their dense vegetation and complex structure. The advent of ICESat-2 and GEDI, two advanced lidar datasets, offers new opportunities for improving canopy height estimation. In this study, we used footprint-level canopy height products from ICESat-2 and GEDI, combined with features extracted from Landsat-8, PALSAR-2, and FABDEM products. The AutoGluon stacking ensemble learning algorithm was employed to construct inversion models, generating 30 m resolution continuous canopy height maps for the tropical forests of Puerto Rico. Accuracy validation was performed using the high-resolution G-LiHT airborne lidar products. Results show that tropical forest canopy height inversion remains challenging, with all models yielding relative root mean square errors (rRMSE) exceeding 0.30. The stacking ensemble model outperformed all base learners, and the GEDI-based map had slightly higher accuracy than the ICESat-2-based map, with RMSE values of 4.81 and 4.99 m, respectively. Both models showed systematic biases, but the GEDI-based model exhibited less underestimation for taller canopies, making it more suitable for biomass estimation. The proposed approach can be applied to other forest ecosystems, enabling fine-resolution canopy height mapping and enhancing forest conservation efforts.

Traffic noise prediction model using GIS and ensemble machine learning: a case study at Universiti Teknologi Malaysia (UTM) Campus.

This study represents a pioneering effort to integrate geographic information systems (GIS) and ensemble machine learning methods to predict noise levels on a university campus. Three ensemble models including random forest (RF), gradient boosting (GB), and extreme gradient boosting (XGB) were developed to predict traffic noise based on data collected over a 4-week period at the Universiti Teknologi Malaysia (UTM) campus. Noise measurements were obtained during peak morning hours (7:30 to 9:30 a.m.) on weekdays within the UTM campus in Johor. Additional predictor variables, including data from the digital elevation model (DEM) and land use, were incorporated to capture the complex nonlinear relationships influencing noise levels. The models were optimized through hyperparameter tuning, resulting in high precision, as evidenced by performance metrics such as the coefficient of determination (R2), mean absolute error (MAE), and mean squared error (MSE). The XGB model emerged as the most accurate, with R2 = 0.96, MAE = 0.9, and MSE = 0.3. Noise maps generated using the inverse distance weighting (IDW) interpolation technique highlighted the spatial distribution of noise levels, classified into five classes considering WHO standards. The findings identified distance from roads, the number of light vehicles, and proximity to green areas as the most significant predictors. However, challenges remain in accurately predicting noise levels associated with other predictors. The outcomes of the study indicate the superior performance of the XGB model compared to the GB and RF models. The study recommends several measures to manage and control noise pollution on the UTM campus, including raising awareness, regulating and enforcing vehicle speed limits, reevaluating land use, installing sound insulation systems, and planting trees and vegetation buffer zones around and within educational buildings.