RESUMEN

Geographical distribution and diversity patterns of bird species are influenced by climate change. The Rouget's rail (Rougetius rougetii) is a ground-dwelling endemic bird species distributed in Ethiopia and Eritrea. It is a near-threatened species menaced by habitat loss, one of the main causes of population declines for bird species. The increasing effects of climate change may further threaten the species' survival. So far, the spatial distribution of this species is not fully documented. With this study, we develop current potential suitable habitat and predict the future habitat shift of R. rougetii based on environmental data such as bioclimatic variables, population density, vegetation cover, and elevation using 10 algorithms. We evaluated the importance of environmental factors in shaping the bird's distribution and how it shifts under climate change scenarios. We used 182 records of R. rougetii from Ethiopia and nine bioclimatic, population density, vegetation cover, and elevation variables to run the 10 model algorithms. Among 10 algorithms, eight were selected for ensembling models according to their predictive abilities. The current suitable habitats for R. rougetii were predicted to cover an area of about 82,000 km2 despite being highly fragmented. The model suggested that temperature seasonality (bio4), elevation, and mean daily air temperatures of the driest quarter (bio9) contributed the most to delimiting suitable areas for this species. R. rougetii is sensitive to climate change associated with elevation, which leads shrinking distribution of suitable areas. The projected spatial and temporal pattern of habitat loss of R. rougetii suggests the importance of climate change mitigation and implementing long-term conservation and management strategies for this threatened endemic bird species.

RESUMEN

Floods are among the natural hazards that have seen a rapid increase in frequency in recent decades. The damage caused by floods, including human and financial losses, poses a serious threat to human life. This study evaluates two machine learning (ML) techniques for flood susceptibility mapping (FSM) in the Gamasyab watershed in Iran. We utilized random forest (RF), support vector machine (SVM), ensemble models, and a geographic information system (GIS) to predict FSM. The application of these models involved 10 effective factors in flooding, as well as 82 flood locations integrated into the GIS. The SVM and RF models were trained and tested, followed by the implementation of resampling techniques (RT) using bootstrap and subsampling methods in three repetitions. The results highlighted the importance of elevation, slope, and precipitation as primary factors influencing flood occurrence. Additionally, the ensemble model outperformed both the RF and SVM models, achieving an area under the curve (AUC) of 0.9, a correlation coefficient (COR) of 0.79, a true skill statistic (TSS) of 0.83, and a standard deviation (SD) of 0.71 in the test phase. The tested models were adapted to available input data to map the FSM across the study watershed. These findings underscore the potential of integrating an ensemble model with GIS as an effective tool for flood susceptibility mapping.

Asunto(s)

RESUMEN

The global trade of non-native pet birds has increased in recent decades, and this has accelerated the introduction of invasive birds in the wild. This study employed ensemble species distribution modelling (eSDM) to assess potential habitat suitability and environmental predictor variables influencing the potential distribution of non-native pet bird species reported lost and sighted in South Africa. We used data and information on lost and found pet birds from previous studies to establish and describe scenarios of how pet birds may transition from captivity to the wild. Our study revealed that models fitted and performed well in predicting the suitability for African grey (Psittacus erithacus), Budgerigar (Melopsittacus undulatus), Cockatiel (Nymphicus hollandicus), Green-cheeked conure (Pyrrhura molinae), Monk parakeet (Myiopsitta monachus), and Rose-ringed parakeet (Psittacula krameri), with the mean weighted AUC and TSS values greater than 0.765. The predicted habitat suitability differed among species, with the suitability threshold indicating that between 61% and 87% of areas were predicted as suitable. Species with greater suitability included the African grey, Cockatiel, and Rose-ringed parakeet, which demonstrated significant overlap between their habitat suitability and reported lost cases. Human footprint, bioclimatic variables, and vegetation indices largely influenced predictive habitat suitability. The pathway scenario showed the key mechanisms driving the transition of pet birds from captivity to the wild, including the role of pet owners, animal rescues, adoption practices, and environmental suitability. Our study found that urban landscapes, which are heavily populated, are at high risk of potential invasion by pet birds. Thus, implementing a thorough surveillance survey is crucial for monitoring and evaluating the establishment potential of pet species not yet reported in the wild.

RESUMEN

BACKGROUND: This study investigated alternative, non-invasive methods for human papillomavirus (HPV) detection in head and neck cancers (HNCs). We compared two approaches: analyzing computed tomography (CT) scans with a Deep Learning (DL) model and using radiomic features extracted from CT images with machine learning (ML) models. METHODS: Fifty patients with histologically confirmed HNC were included. We first trained a modified ResNet-18 DL model on CT data to predict HPV status. Next, radiomic features were extracted from manually segmented regions of interest near the oropharynx and used to train four ML models (K-Nearest Neighbors, logistic regression, decision tree, random forest) for the same purpose. RESULTS: The CT-based model achieved the highest accuracy (90%) in classifying HPV status. Among the ML models, K-Nearest Neighbors performed best (80% accuracy). Weighted Ensemble methods combining the CT-based model with each ML model resulted in moderate accuracy improvements (70-90%). CONCLUSIONS: Our findings suggest that CT scans analyzed by DL models hold promise for non-invasive HPV detection in HNC. Radiomic features, while less accurate in this study, offer a complementary approach. Future research should explore larger datasets and investigate the potential of combining DL and radiomic techniques.

RESUMEN

Driving while drowsy poses significant risks, including reduced cognitive function and the potential for accidents, which can lead to severe consequences such as trauma, economic losses, injuries, or death. The use of artificial intelligence can enable effective detection of driver drowsiness, helping to prevent accidents and enhance driver performance. This research aims to address the crucial need for real-time and accurate drowsiness detection to mitigate the impact of fatigue-related accidents. Leveraging ultra-wideband radar data collected over five minutes, the dataset was segmented into one-minute chunks and transformed into grayscale images. Spatial features are retrieved from the images using a two-dimensional Convolutional Neural Network. Following that, these features were used to train and test multiple machine learning classifiers. The ensemble classifier RF-XGB-SVM, which combines Random Forest, XGBoost, and Support Vector Machine using a hard voting criterion, performed admirably with an accuracy of 96.6%. Additionally, the proposed approach was validated with a robust k-fold score of 97% and a standard deviation of 0.018, demonstrating significant results. The dataset is augmented using Generative Adversarial Networks, resulting in improved accuracies for all models. Among them, the RF-XGB-SVM model outperformed the rest with an accuracy score of 99.58%.

Asunto(s)

RESUMEN

This paper investigates the application of ensemble learning techniques, specifically meta-learning, in intrusion detection systems (IDS) for the Internet of Medical Things (IoMT). It underscores the existing challenges posed by the heterogeneous and dynamic nature of IoMT environments, which necessitate adaptive, robust security solutions. By harnessing meta-learning alongside various ensemble strategies such as stacking and bagging, the paper aims to refine IDS mechanisms to effectively counter evolving cyber threats. The study proposes a performance-driven weighted meta-learning technique for dynamic assignment of voting weights to classifiers based on accuracy, loss, and confidence levels. This approach significantly enhances the intrusion detection capabilities for the IoMT by dynamically optimizing ensemble IDS models. Extensive experiments demonstrate the proposed model's superior performance in terms of accuracy, detection rate, F1 score, and false positive rate compared to existing models, particularly when analyzing various sizes of input features. The findings highlight the potential of integrating meta-learning in ensemble-based IDS to enhance the security and integrity of IoMT networks, suggesting avenues for future research to further advance IDS performance in protecting sensitive medical data and IoT infrastructures.

RESUMEN

Equipment downtime resulting from maintenance in various sectors around the globe has become a major concern. The effectiveness of conventional reactive maintenance methods in addressing interruptions and enhancing operational efficiency has become inadequate. Therefore, acknowledging the constraints associated with reactive maintenance and the growing need for proactive approaches to proactively detect possible breakdowns is necessary. The need for optimisation of asset management and reduction of costly downtime emerges from the demand for industries. The work highlights the use of Internet of Things (IoT)-enabled Predictive Maintenance (PdM) as a revolutionary strategy across many sectors. This article presents a picture of a future in which the use of IoT technology and sophisticated analytics will enable the prediction and proactive mitigation of probable equipment failures. This literature study has great importance as it thoroughly explores the complex steps and techniques necessary for the development and implementation of efficient PdM solutions. The study offers useful insights into the optimisation of maintenance methods and the enhancement of operational efficiency by analysing current information and approaches. The article outlines essential stages in the application of PdM, encompassing underlying design factors, data preparation, feature selection, and decision modelling. Additionally, the study discusses a range of ML models and methodologies for monitoring conditions. In order to enhance maintenance plans, it is necessary to prioritise ongoing study and improvement in the field of PdM. The potential for boosting PdM skills and guaranteeing the competitiveness of companies in the global economy is significant through the incorporation of IoT, Artificial Intelligence (AI), and advanced analytics.

RESUMEN

INTRODUCTION: The extraction decision significantly affects the treatment process and outcome. Therefore, it is crucial to make this decision with a more objective and standardized method. The objectives of this study were (1) to identify the best-performing model among seven machine learning (ML) models, which will standardize the extraction decision and serve as a guide for inexperienced clinicians, and (2) to determine the important variables for the extraction decision. METHODS: This study included 1000 patients who received orthodontic treatment with or without extraction (500 extraction and 500 non-extraction). The success criteria of the study were the decisions made by the four experienced orthodontists. Seven ML models were trained using 36 variables; including demographic information, cephalometric and model measurements. First, the extraction decision was performed, and then the extraction type was identified. Accuracy and area under the curve (AUC) of the receiver operating characteristics (ROC) curve were used to measure the success of ML models. RESULTS: The Stacking Classifier model, which consists of Gradient Boosted Trees, Support Vector Machine, and Random Forest models, showed the highest performance in extraction decision with 91.2% AUC. The most important features determining extraction decision were maxillary and mandibular arch length discrepancy, Wits Appraisal, and ANS-Me length. Likewise, the Stacking Classifier showed the highest performance with 76.3% accuracy in extraction type decisions. The most important variables for the extraction type decision were mandibular arch length discrepancy, Class I molar relationship, cephalometric overbite, Wits Appraisal, and L1-NB distance. CONCLUSION: The Stacking Classifier model exhibited the best performance for the extraction decision. While ML models showed a high performance in extraction decision, they could not able to achieve the same level of performance in extraction type decision.

RESUMEN

BACKGROUND: Lymphatic filariasis (LF) is a neglected tropical disease targeted for elimination as a public health problem by 2030. Although mass treatments have led to huge reductions in LF prevalence, some countries or regions may find it difficult to achieve elimination by 2030 owing to various factors, including local differences in transmission. Subnational projections of intervention impact are a useful tool in understanding these dynamics, but correctly characterizing their uncertainty is challenging. METHODS: We developed a computationally feasible framework for providing subnational projections for LF across 44 sub-Saharan African countries using ensemble models, guided by historical control data, to allow assessment of the role of subnational heterogeneities in global goal achievement. Projected scenarios include ongoing annual treatment from 2018 to 2030, enhanced coverage, and biannual treatment. RESULTS: Our projections suggest that progress is likely to continue well. However, highly endemic locations currently deploying strategies with the lower World Health Organization recommended coverage (65%) and frequency (annual) are expected to have slow decreases in prevalence. Increasing intervention frequency or coverage can accelerate progress by up to 5 or 6 years, respectively. CONCLUSIONS: While projections based on baseline data have limitations, our methodological advancements provide assessments of potential bottlenecks for the global goals for LF arising from subnational heterogeneities. In particular, areas with high baseline prevalence may face challenges in achieving the 2030 goals, extending the "tail" of interventions. Enhancing intervention frequency and/or coverage will accelerate progress. Our approach facilitates preimplementation assessments of the impact of local interventions and is applicable to other regions and neglected tropical diseases.

Asunto(s)

RESUMEN

Cardiovascular disease (CVD) stands as the leading global cause of mortality, and coronary artery disease (CAD) has the highest prevalence, contributing to 42% of these fatalities. Recognizing the constraints inherent in the anatomical assessment of CAD, Fractional Flow Reserve (FFR) has emerged as a pivotal functional diagnostic metric. Herein, we assess the potential of employing an ensemble approach with deep neural networks (DNN) to predict invasively measured Fractional Flow Reserve (FFR) using raw anatomical data extracted from both optical coherence tomography (OCT) and X-ray coronary angiography (XA). In this study, we used a challenging dataset, with 46% of the lesions falling within the FFR range of 0.75 to 0.85. Despite this complexity, our model achieved an accuracy of 84.3%, demonstrating a sensitivity of 87.5% and a specificity of 81.4%. Our results demonstrate that incorporating both OCT and XA signals, co-registered, as inputs for the DNN model leads to an important increase in overall accuracy.

Asunto(s)

RESUMEN

This study addresses the challenges associated with emergency department (ED) overcrowding and emphasizes the need for efficient risk stratification tools to identify high-risk patients for early intervention. While several scoring systems, often based on logistic regression (LR) models, have been proposed to indicate patient illness severity, this study aims to compare the predictive performance of ensemble learning (EL) models with LR for in-hospital mortality in the ED. A cross-sectional single-center study was conducted at the ED of Imam Reza Hospital in northeast Iran from March 2016 to March 2017. The study included adult patients with one to three levels of emergency severity index. EL models using Bagging, AdaBoost, random forests (RF), Stacking and extreme gradient boosting (XGB) algorithms, along with an LR model, were constructed. The training and validation visits from the ED were randomly divided into 80% and 20%, respectively. After training the proposed models using tenfold cross-validation, their predictive performance was evaluated. Model performance was compared using the Brier score (BS), The area under the receiver operating characteristics curve (AUROC), The area and precision-recall curve (AUCPR), Hosmer-Lemeshow (H-L) goodness-of-fit test, precision, sensitivity, accuracy, F1-score, and Matthews correlation coefficient (MCC). The study included 2025 unique patients admitted to the hospital's ED, with a total percentage of hospital deaths at approximately 19%. In the training group and the validation group, 274 of 1476 (18.6%) and 152 of 728 (20.8%) patients died during hospitalization, respectively. According to the evaluation of the presented framework, EL models, particularly Bagging, predicted in-hospital mortality with the highest AUROC (0.839, CI (0.802-0.875)) and AUCPR = 0.64 comparable in terms of discrimination power with LR (AUROC (0.826, CI (0.787-0.864)) and AUCPR = 0.61). XGB achieved the highest precision (0.83), sensitivity (0.831), accuracy (0.842), F1-score (0.833), and the highest MCC (0.48). Additionally, the most accurate models in the unbalanced dataset belonged to RF with the lowest BS (0.128). Although all studied models overestimate mortality risk and have insufficient calibration (P > 0.05), stacking demonstrated relatively good agreement between predicted and actual mortality. EL models are not superior to LR in predicting in-hospital mortality in the ED. Both EL and LR models can be considered as screening tools to identify patients at risk of mortality.

Asunto(s)

RESUMEN

Adsorption is one of the most promising wastewater treatment methods due to its simplicity and efficacy at ambient temperature and pressure. However, the technical and economic feasibility of this process largely depends on the performance of the utilized adsorbents. In this study, a promising adsorbent made of polyethyleneimine, graphene oxide (GO), bentonite, and MgFeAl-layered triple hydroxide (MgFeAl-LTH) has been synthesized and characterized. The results revealed that the synthesized nanocomposite (abbreviated as PGB-LTH) possesses good porosity and crystallinity. The adsorption performance of the PGB-LTH nanocomposite towards two harmful water pollutants (i.e., methyl orange (MO) and crystal violet (CV)) was investigated, and the results revealed that the nanocomposite outperforms its parental materials (i.e., GO, bentonite, and MgFeAl-LTH). The maximum adsorption capacity (qmax) of MO and CV onto the nanocomposite could reach 1666.7 and 1250.0 mg/g, respectively, as predicted using the Langmuir adsorption isotherm. Additionally, the PGB-LTH nanocomposite is highly reusable with an insignificant decline in performance upon repetitive use. In terms of thermodynamics, MO adsorption onto the nanocomposite is exothermic while CV adsorption is endothermic despite that both dyes adsorb spontaneously as revealed by the negative values of the Gibbs free energy change at all the examined temperatures. The generated adsorption data were utilized for constructing and assessing ensemble meta-machine learning techniques aimed at cost-effective simulation and prediction of the proposed adsorption method. Bagging and boosting methods were developed and evaluated intensively using the obtained adsorption data. The Extra Trees model achieved promising results as evidenced by the high correlation coefficient of 99% as well as low computed RMSE and MAE errors of 11.42 and 5.11, respectively, during the testing phase. These results demonstrate the model strong capability to effectively simulate and predict the adsorption process in question.

Asunto(s)

RESUMEN

In the contemporary landscape of healthcare, the early and accurate prediction of diabetes has garnered paramount importance, especially in the wake of the COVID-19 pandemic where individuals with diabetes exhibit increased vulnerability. This research embarked on a mission to enhance diabetes prediction by employing state-of-the-art machine learning techniques. Initial evaluations highlighted the Support Vector Machines (SVM) classifier as a promising candidate with an accuracy of 76.62%. To further optimize predictions, the study delved into advanced feature engineering techniques, generating interaction and polynomial features that unearthed hidden patterns in the data. Subsequent correlation analyses, visualized through heatmaps, revealed significant correlations, especially with attributes like Glucose. By integrating the strengths of Decision Trees, Gradient Boosting, and SVM in an ensemble model, we achieved an accuracy of 93.2%, showcasing the potential of harmonizing diverse algorithms. This research offers a robust blueprint for diabetes prediction, holding profound implications for early diagnosis, personalized treatments, and preventive care in the context of global health challenges and with the goal of increasing life expectancy.

Asunto(s)

RESUMEN

Convolutional neural networks have achieved great success in computer vision, but incorrect predictions would be output when applying intended perturbations on original input. These human-indistinguishable replicas are called adversarial examples, which on this feature can be used to evaluate network robustness and security. White-box attack success rate is considerable, when already knowing network structure and parameters. But in a black-box attack, the adversarial examples success rate is relatively low and the transferability remains to be improved. This article refers to model augmentation which is derived from data augmentation in training generalizable neural networks, and proposes resizing invariance method. The proposed method introduces improved resizing transformation to achieve model augmentation. In addition, ensemble models are used to generate more transferable adversarial examples. Extensive experiments verify the better performance of this method in comparison to other baseline methods including the original model augmentation method, and the black-box attack success rate is improved on both the normal models and defense models.

RESUMEN

BACKGROUND: With rising incidence of skin cancer and relatively increased mortality rates, an improved diagnosis of such a potentially fatal disease is of vital importance. Although frequently curable, it nevertheless places a considerable burden upon healthcare systems. Among the various types of skin cancers, non-melanoma skin cancer is most prevalent. Despite such prevalence and its associated cost, scant proof concerning the diagnostic accuracy via Artificial Intelligence (AI) for non-melanoma skin cancer exists. This study meta-analyzes the diagnostic test accuracy of AI used to diagnose non-melanoma forms of skin cancer, and it identifies potential covariates that account for heterogeneity between extant studies. METHODS: Various electronic databases (Scopus, PubMed, ScienceDirect, SpringerLink, and Dimensions) were examined to discern eligible studies beginning from March 2022. Those AI studies predictive of non-melanoma skin cancer were included. Summary estimates of sensitivity, specificity, and area under receiver operating characteristic curves were used to evaluate diagnostic accuracy. The revised Quality Assessment of Diagnostic Studies served to assess any risk of bias. RESULTS: A literature search produced 39 eligible articles for meta-analysis. The summary sensitivity, specificity, and area under receiver operating characteristic curve of AI for diagnosing non-melanoma skin cancer was 0.78, 0.98, & 0.97, respectively. Skin cancer typology, data sources, cross validation, ensemble models, types of techniques, pre-trained models, and image augmentation became significant covariates accounting for heterogeneity in terms of both sensitivity and/or specificity. CONCLUSIONS: Meta-analysis results revealed that AI is predictive of non-melanoma with an acceptable performance, but sensitivity may become improved. Further, ensemble models and pre-trained models are employable to improve true positive rating.

Asunto(s)

RESUMEN

Introduction: Since the advent of medical education systems, managing high-stakes exams has been a top priority and challenge for all policymakers. However, considering machine learning (ML) techniques as a replacement for medical licensing examinations, particularly during crises such as the COVID-19 outbreak, could be an effective solution. This study uses ML models to develop a framework for predicting medical students' performance on high-stakes exams, such as the Comprehensive Medical Basic Sciences Examination (CMBSE). Material and methods: Prediction of students' status and score on high-stakes examinations faces several challenges, including an imbalanced number of failing and passing students, a large number of heterogeneous and complex features, and the need to identify at-risk and top-performing students. In this study, two major categories of ML approaches are compared: first, classic models (logistic regression (LR), support vector machine (SVM), and k-nearest neighbors (KNN)), and second, ensemble models (voting, bagging (BG), random forests (RF), adaptive boosting (ADA), extreme gradient boosting (XGB), and stacking). Results: To evaluate the models' discrimination ability, they are assessed using a real dataset containing information on medical students over a five-year period (n = 1005). The findings indicate that ensemble ML models demonstrate optimal performance in predicting CMBSE status (RF and stacking). Similarly, among the classic regressors, LR exhibited the highest root-mean-square deviation (RMSD) (0.134) and coefficient of determination (R2) (0.62), whereas the RF model had the highest RMSD (0.077) and R2 (0.80) overall. Furthermore, Anatomical Sciences, Biochemistry, Parasitology, and Entomology grade point average (GPA) and grades demonstrated the strongest positive correlation with the outcomes. Conclusion: Comparing classic and ensemble ML models revealed that ensemble models are superior to classic models. Therefore, the presented framework could be considered a suitable alternative for the CMBSE and other comparable medical licensing examinations.

RESUMEN

In this research, we demonstrate a Deep Convolutional Neural Network-based classification model for the detection of monkeypox. Monkeypox can be difficult to diagnose clinically in its early stages since it resembles both chickenpox and measles in symptoms. The early diagnosis of monkeypox helps doctors cure it more quickly. Therefore, pre-trained models are frequently used in the diagnosis of monkeypox, because the manual analysis of a large number of images is labor-intensive and prone to inaccuracy. Therefore, finding the monkeypox virus requires an automated process. The large layer count of convolutional neural network (CNN) architectures enables them to successfully conceptualize the features on their own, thereby contributing to better performance in image classification. The scientific community has recently articulated significant attention in employing artificial intelligence (AI) to diagnose monkeypox from digital skin images due primarily to AI's success in COVID-19 identification. The VGG16, VGG19, ResNet50, ResNet101, DenseNet201, and AlexNet models were used in our proposed method to classify patients with monkeypox symptoms with other diseases of a similar kind (chickenpox, measles, and normal). The majority of images in our research are collected from publicly available datasets. This study suggests an adaptive k-means clustering image segmentation technique that delivers precise segmentation results with straightforward operation. Our preliminary computational findings reveal that the proposed model could accurately detect patients with monkeypox. The best overall accuracy achieved by ResNet101 is 94.25%, with an AUC of 98.59%. Additionally, we describe the categorization of our model utilizing feature extraction using Local Interpretable Model-Agnostic Explanations (LIME), which provides a more in-depth understanding of particular properties that distinguish the monkeypox virus.

RESUMEN

The medical and healthcare domains require automatic diagnosis systems (ADS) for the identification of health problems with technological advancements. Biomedical imaging is one of the techniques used in computer-aided diagnosis systems. Ophthalmologists examine fundus images (FI) to detect and classify stages of diabetic retinopathy (DR). DR is a chronic disease that appears in patients with long-term diabetes. Unattained patients can lead to severe conditions of DR, such as retinal eye detachments. Therefore, early detection and classification of DR are crucial to ward off advanced stages of DR and preserve the vision. Data diversity in an ensemble model refers to the use of multiple models trained on different subsets of data to improve the ensemble's overall performance. In the context of an ensemble model based on a convolutional neural network (CNN) for diabetic retinopathy, this could involve training multiple CNNs on various subsets of retinal images, including images from different patients or those captured using distinct imaging techniques. By combining the predictions of these multiple models, the ensemble model can potentially make more accurate predictions than a single prediction. In this paper, an ensemble model (EM) of three CNN models is proposed for limited and imbalanced DR data using data diversity. Detecting the Class 1 stage of DR is important to control this fatal disease in time. CNN-based EM is incorporated to classify the five classes of DR while giving attention to the early stage, i.e., Class 1. Furthermore, data diversity is created by applying various augmentation and generation techniques with affine transformation. Compared to the single model and other existing work, the proposed EM has achieved better multi-class classification accuracy, precision, sensitivity, and specificity of 91.06%, 91.00%, 95.01%, and 98.38%, respectively.

RESUMEN

Lake surface water temperature is one of the most important physical and ecological indices of lakes, which has frequently been used as the indicator to evaluate the impact of climate change on lakes. Knowing the dynamics of lake surface water temperature is thus of great significance. The past decades have witnessed the development of different modeling tools to forecast lake surface water temperature, yet, simple models with fewer input variables, while maintaining high forecasting accuracy are scarce. Impact of forecast horizons on model performance has seldom been investigated. To fill the gap, in this study, a novel machine learning algorithm by stacking multilayer perceptron and random forest (MLP-RF) was employed to forecast daily lake surface water temperature using daily air temperature as the exogenous input variable, with the Bayesian Optimization procedure applied for tuning the hyperparameters. Prediction models were developed using long-term observed data from eight Polish lakes. The MLP-RF stacked model showed very good forecasting capabilities for all lakes and forecast horizons, far better than shallow multilayer perceptron neural network, a model coupling wavelet transform and multilayer perceptron neural network, non-linear regression and air2water models. A reduction in model performance was observed as the forecast horizon increased. However, the model also performs well with a forecast horizon of several days (e.g., 7 days ahead, testing stage: R2 - [0.932, 0.990], RMSE °C - [0.77, 1.83], MAE °C - [0.55, 1.38]). In addition, the MLP-RF stacked model has proven to be reliable for both intermediate temperatures and minimum and maximum peaks. The model proposed in this study will be useful to the scientific community in predicting lake surface water temperature, thus contributing to studies on such sensitive aquatic ecosystems as lakes.

Asunto(s)

RESUMEN

Gaussian graphical models (GGMs) are a popular form of network model in which nodes represent features in multivariate normal data and edges reflect conditional dependencies between these features. GGM estimation is an active area of research. Currently available tools for GGM estimation require investigators to make several choices regarding algorithms, scoring criteria, and tuning parameters. An estimated GGM may be highly sensitive to these choices, and the accuracy of each method can vary based on structural characteristics of the network such as topology, degree distribution, and density. Because these characteristics are a priori unknown, it is not straightforward to establish universal guidelines for choosing a GGM estimation method. We address this problem by introducing SpiderLearner, an ensemble method that constructs a consensus network from multiple estimated GGMs. Given a set of candidate methods, SpiderLearner estimates the optimal convex combination of results from each method using a likelihood-based loss function. K $$ K $$ -fold cross-validation is applied in this process, reducing the risk of overfitting. In simulations, SpiderLearner performs better than or comparably to the best candidate methods according to a variety of metrics, including relative Frobenius norm and out-of-sample likelihood. We apply SpiderLearner to publicly available ovarian cancer gene expression data including 2013 participants from 13 diverse studies, demonstrating our tool's potential to identify biomarkers of complex disease. SpiderLearner is implemented as flexible, extensible, open-source code in the R package ensembleGGM at https://github.com/katehoffshutta/ensembleGGM.

Asunto(s)