A Neural Network-Based Predictive Model for Forest Fire Management in Michoacán

In the field of oil and gas well perforation, it is imperative to accurately forecast the casing entry hole diameter under full downhole conditions. Precise prediction of the casing entry hole diameter enhances the design of both conventional and limited entry hydraulic fracturing, mitigates the risk of proppant screenout, reduces skin factors attributable to perforation, guarantees the presence of sufficient flow areas for the effective pumping of cement during a squeeze operation, and reduces issues related to sand production. Implementing machine learning and deep learning models yields immediate and precise estimations of entry hole diameter, thereby facilitating the attainment of these objectives. The principal aim of this research is to develop sophisticated machine learning-based models proficient in predicting entry hole diameter under full downhole conditions. Ten machine learning and deep learning models have been developed utilizing readily available parameters routinely gathered during perforation operations, including perforation depth, rock density, shot phasing, shot density, fracture gradient, reservoir unconfined compressive strength, casing elastic limit, casing nominal weight, casing outer diameter, and gun diameter as input variables. These models are trained by utilizing actual casing entry hole diameter data acquired from deployed downhole cameras, which serve as the output for the X’ models. A comprehensive dataset from 53 wells has been utilized to meticulously develop and fine-tune various machine learning algorithms. These include Gradient Boosting, Linear Regression, Stochastic Gradient Descent, AdaBoost, Decision Trees, Random Forest, K-Nearest Neighbor, neural network, and Support Vector Machines. The results of the most effective machine learning models, specifically Gradient Boosting, Random Forest, AdaBoost, neural network (L-BFGS), and neural network (Adam), reveal exceptionally low values of mean absolute percent error (MAPE), root mean square error (RMSE), and mean squared error (MSE) in comparison to actual measurements of entry hole diameter. The recorded MAPE values are 4.6%, 4.4%, 4.7%, 4.9%, and 6.3%, with corresponding RMSE values of 0.057, 0.057, 0.058, 0.065, and 0.089, and MSE values of 0.003, 0.003, 0.003, 0.004, and 0.008, respectively. These low MAPE, RMSE, and MSE values verify the remarkably high accuracy of the generated models. This paper offers novel insights by demonstrating the improvements achieved in ongoing perforation operations through the application of a machine learning model for predicting entry hole diameter. The utilization of machine learning models presents a more accurate, expedient, real-time, and economically viable alternative to empirical models and deployed downhole cameras. Additionally, these machine learning models excel in accommodating a broad spectrum of guns, well completions, and reservoir parameters, a challenge that a singular empirical model struggled to address.

Electrical submersible pumps (ESPs) are among the most common artificial lift techniques in highly productive oil wells. The ESP failures are extremely costly to the producers and must be minimized. This study proposes a hybrid approach utilizing multi-class classification machine learning (ML) models to identify various specific failure modes (SFMs) of an ESP. A comprehensive dataset and various ML algorithms are utilized, considering the physics of fluid flow through the ESP. The ML models are based on field data gathered from the surface and downhole ESP monitoring equipment over five years of production of 10 wells. The dataset includes the failure cause, duration of downtime, the corresponding high-frequency (per minute) pump data, and well-production data. The prediction periods of 3 hours to 7 days before the failure are evaluated to minimize false alarms and predict the true events. Four modeling designs are used to handle the data and predict ESP failure. These designs differ in the input parameters used for the model and signify the effect of including the physical parameters in failure prediction. Several ML models are tested and evaluated using precision, recall, and F1-score performance measures. The K-Nearest Neighbor (KNN) model outperforms the other algorithms in forecasting ESP failures. Some other tested models are Random Forest (RF), Decision Tree (DT), Multilayer Perceptron (MLP) Neural Network, etc. According to the data, most ESP operational failures are characterized as electrical failures. The ML models show similarly good performances with high true prediction rates in predicting ESP failures for all the tested designs. The design that integrates the effects of gas presence and pump efficiency while minimizing the number of input variables is suggested for general use. Increasing the prediction period up to 3 days results in a negligible drop in the model’s performance, showing that the model can predict ESP failures accurately three days before their occurrences. However, the forecasts show increases in missed failures and false alarms for prediction periods of more than three days, making three days the selected prediction period. These ML models will aid operators in avoiding undesirable events, reducing downtime, and extending the lifespan of ESPs. ESP failures are unanticipated but common occurrences in oil and gas wells. It is necessary to detect the onset of failures early and prevent excessive downtime. This study’s model allows engineers to detect failures early, diagnose potential causes, and propose preventive actions. It is crucial in transitioning from a reactive event-based to proactive and predictive maintenance of artificial lift operations.

Background and Aims Acute kidney injury (AKI) has a substantial impact on global disease burden of Chronic Kidney Disease. To assist physicians with the timely diagnosis of AKI, several prognostic models have been developed to improve early recognition across various patient populations with varying degrees of predictive performance. In the prediction of AKI, machine learning (ML) techniques have been demonstrated to improve on the predictive ability of existing models that rely on more conventional statistical methods. ML is a broad term which refers to various types of models: Parametric models, such as linear or logistic regression use a pre-specified model form which is believed to fit the data, and its parameters are estimated. Non-parametric models, such as decision trees, random forests, and neural networks may have varying complexity (e.g. the depth of a classification tree model) based on the data. Deep learning neural network models exploit temporal or spatial arrangements in the data to deal with complex predictors. Given the rapid growth and development of ML methods and models for AKI prediction over the past years, in this systematic review, we aim to appraise the current state-of-the-art regarding ML models for the prediction of AKI. To this end, we focus on model performance, model development methods, model evaluation, and methodological limitations. Method We searched the PubMed and ArXiv digital libraries, and selected studies that develop or validate an AKI-related multivariable ML prediction model. We extracted data using a data extraction form based on the TRIPOD (transparent reporting of a multivariable prediction model for individual prognosis or diagnosis) and CHARMS (critical appraisal and data extraction for systematic reviews of prediction modelling studies) checklists. Results Overall, 2,875 titles were screened and thirty-four studies were included. Of those, thirteen studies focussed on intensive care, for which the US derived MIMIC dataset was commonly used; thirty-one studies both developed and validated a model; twenty-one studies used single-centre data. Non-parametric ML methods were used more often than regression and deep learning. Random forests was the most popular method, and often performed best in model comparisons. Deep learning was typically used (and also effective) when complex features were included (e.g., with text or time series). Internal validation was often applied, and the performance of ML models was usually compared against logistic regression. However, the simple training/test split was often used, which does not account for the variability of the training and test samples. Calibration, external validation, and interpretability of results were rarely considered. Comparisons of model performance against medical scores or clinicians were also rare. Reproducibility was limited, as data and code were usually unavailable. Conclusion There is an increasing number of ML models for AKI, which are mostly developed in the intensive care environment largely due to the availability of the MIMIC dataset. Most studies are single-centre, and lack a prospective design. More complex models based on deep learning are emerging, with the potential to improve predictions for complex data, such as time-series, but with the disadvantage of being less interpretable. Future studies should pay attention to using calibration measures, external validation, and on improving model interpretability, in order to improve uptake in clinical practice. Finally, sharing data and code could improve reproducibility of study findings.

Machine learning (ML) and deep learning (DL) models can play an important role when it comes to modelling complicated processes. Such capability is necessary for hydrological and climate-related applications. Generally, ML models utilize precipitation and temperature time series of a basin as input to develop a lumped rainfall-runoff model to simulate streamflow at the basin outlet. However, when it is divided into several sub-basins, Graph Neural Networks (GNN) can consider each sub-basin as a node and link them together using a connectivity matrix to account for spatial variations of hydroclimatic variables. In this study, GNN and various ML models with different types of architecture, ranging from neural networks, tree-based structure, and gradient boosting, were exploited for daily streamflow simulation over different case studies. For each case study, the basin was divided into a few sub-basins for which daily precipitation and temperature data were aggregated and used as input. For training GNN, the connection matrix of sub-basins was also used as input. Basically, 75% of historical records were utilized to train GNN and different ML models, e.g., artificial neural networks, support vector machine, decision tree, random forest, eXtreme Gradient Boosting (XGBoost), Light Gradient-Boosting Machine (LightGBM), and Category Boosting (CatBoost), while the rest was used for testing. Streamflow simulation was conducted with/without considering seasonality impact and lag times. The obtained results clearly demonstrate that considering seasonality and time lags can enhance accuracy of streamflow predictions based on Kling–Gupta efficiency (KGE). Furthermore, GNN with seasonality impact and time lags achieved promising results across different case studies with KGE>0.85 for training and KGE>0.59 for testing data, respectively. Among ML models, boosting models, e.g., LightGBM and XGBoost, performed slightly better than other ML models. for Finally, this comparative analysis provides valuable insights for ML/DL applications in climate change impact assessments.Acknowledgements: This research work was carried out as part of the TRANSCEND project with funding received from the European Union Horizon Europe Research and Innovation Programme under Grant Agreement No. 10108411.

Machine learning approaches for formation matrix volume prediction from well logs: Insights and lessons learned

In recent years, a significant amount of research has focused on analyzing the effectiveness of machine learning (ML) models for malware detection. These approaches have ranged from methods such as decision trees and clustering to more complex approaches like support vector machine (SVM) and deep neural networks. In particular, neural networks have proven to be very effective in detecting complex and advanced malware. This, however, comes with a caveat. Neural networks are notoriously complex. Therefore, the decisions that they make are often just accepted without questioning why the model made that specific decision. The black box characteristic of neural networks has challenged researchers to explore methods to explain black-box models such as SVM and neural networks and their decision-making process. Transparency and explainability give the experts and malware analysts assurance and trustworthiness about the ML models’ decisions. In addition, it helps in generating comprehensive reports that can be used to enhance cyber threat intelligence sharing. As such, this much needed analysis drives our work in this paper to explore the explainability and interpretability of ML models in the field of online malware detection. In this paper, we used the Shapley Additive exPlanations (SHAP) explainability technique to achieve efficient performance in interpreting the outcome of different ML models such as SVM Linear, SVM-RBF (Radial Basis Function), Random Forest (RF), Feed-Forward Neural Net (FFNN), and Convolutional Neural Network (CNN) models trained on an online malware dataset. To explain the output of these models, explainability techniques such as KernalSHAP, TreeSHAP, and DeepSHAP are applied to the obtained results.

Forecasting rainfall, crucial for agriculture, water management, and disaster preparedness, presents significant challenges due to intricate relationships often missed by conventional statistical methods. Hence, machine learning (ML) models offer promising alternatives, enhancing the precision and dependability of rainfall predictions. This paper presents a comprehensive comparison of various ML models with diverse model structures and regularization strategies for rainfall prediction in urban metropolitan cities. The results show that the random forest model, and gradient boosting model outperform the other models such as logistic regression, support vector machine (SVM), decision tree, K-nearest neighbor (KNN), Naive Bayes, linear SVM, and neural network in terms of accuracy. Validation accuracies of 75%, 77%, 68%, 78%, 76%, 78%, 74%, 75% and 76% were achieved for logistic regression, SVM, decision tree, random forest model, KNN, gradient boosting model, Naive Bayes, linear SVM, and neural network, respectively. The choice of ML models for rainfall prediction should consider the characteristics of the data, e.g. a lag feature for 20 days was employed that uses previous time steps to predict the next time step. The paper concludes that ML models, especially the random forest and gradient boosting models are powerful and robust tools for rainfall prediction in urban metropolitan cities. Key Words: rainfall prediction, wind information, machine learning, gradient boosting, random forest

To explore the value of machine learning (ML) models based on contrast-enhanced cone-beam breast computed tomography (CE-CBBCT) radiomics features for the preoperative prediction of human epidermal growth factor receptor 2 (HER2)-low expression breast cancer (BC). Fifty-six patients with HER2-negative invasive BC who underwent preoperative CE-CBBCT were prospectively analyzed. Patients were randomly divided into training and validation cohorts at approximately 7:3. A total of 1046 quantitative radiomic features were extracted from CE-CBBCT images and normalized using z-scores. The Pearson correlation coefficient and recursive feature elimination were used to identify the optimal features. Six ML models were constructed based on the selected features: linear discriminant analysis (LDA), random forest (RF), support vector machine (SVM), logistic regression (LR), AdaBoost (AB), and decision tree (DT). To evaluate the performance of these models, receiver operating characteristic curves and area under the curve (AUC) were used. Seven features were selected as the optimal features for constructing the ML models. In the training cohort, the AUC values for SVM, LDA, RF, LR, AB, and DT were 0.984, 0.981, 1.000, 0.970, 1.000, and 1.000, respectively. In the validation cohort, the AUC values for the SVM, LDA, RF, LR, AB, and DT were 0.859, 0.880, 0.781, 0.880, 0.750, and 0.713, respectively. Among all ML models, the LDA and LR models demonstrated the best performance. The DeLong test showed that there were no significant differences among the receiver operating characteristic curves in all ML models in the training cohort (P > .05); however, in the validation cohort, the DeLong test showed that the differences between the AUCs of LDA and RF, AB, and DT were statistically significant (P = .037, .003, .046). The AUCs of LR and RF, AB, and DT were statistically significant (P = .023, .005, .030). Nevertheless, no statistically significant differences were observed when compared to the other ML models. ML models based on CE-CBBCT radiomics features achieved excellent performance in the preoperative prediction of HER2-low BC and could potentially serve as an effective tool to assist in precise and personalized targeted therapy.

In the current paper we assess different machine learning (ML) models and hybrid geostatistical methods in the prediction of soil pH using digital elevation model derivates (environmental covariates) and co-located soil parameters (soil covariates). The study was located in the area of Grevena, Greece, where 266 disturbed soil samples were collected from randomly selected locations and analyzed in the laboratory of the Soil and Water Resources Institute. The different models that were assessed were random forests (RF), random forests kriging (RFK), gradient boosting (GB), gradient boosting kriging (GBK), neural networks (NN), and neural networks kriging (NNK) and finally, multiple linear regression (MLR), ordinary kriging (OK), and regression kriging (RK) that although they are not ML models, they were used for comparison reasons. Both the GB and RF models presented the best results in the study, with NN a close second. The introduction of OK to the ML models’ residuals did not have a major impact. Classical geostatistical or hybrid geostatistical methods without ML (OK, MLR, and RK) exhibited worse prediction accuracy compared to the models that included ML. Furthermore, different implementations (methods and packages) of the same ML models were also assessed. Regarding RF and GB, the different implementations that were applied (ranger-ranger, randomForest-rf, xgboost-xgbTree, xgboost-xgbDART) led to similar results, whereas in NN, the differences between the implementations used (nnet-nnet and nnet-avNNet) were more distinct. Finally, ML models tuned through a random search optimization method were compared with the same ML models with their default values. The results showed that the predictions were improved by the optimization process only where the ML algorithms demanded a large number of hyperparameters that needed tuning and there was a significant difference between the default values and the optimized ones, like in the case of GB and NN, but not in RF. In general, the current study concluded that although RF and GB presented approximately the same prediction accuracy, RF had more consistent results, regardless of different packages, different hyperparameter selection methods, or even the inclusion of OK in the ML models’ residuals.

BackgroundMild Cognitive Impairment (MCI) is a critical transitional stage between normal aging and Alzheimer’s disease, and its early identification is essential for delaying disease progression.MethodsThis study, based on data from the 2020 China Health and Retirement Longitudinal Study (CHARLS), focuses on older adults with functional disability as the target population. LASSO regression, combined with univariable and multivariable logistic regression, was employed to select feature variables for predictive modeling. Seven machine learning algorithms, including logistic regression, decision tree, random forest, support vector machine, gradient boosting decision tree, k-nearest neighbors, and neural network, were used to develop predictive models. Model performance was evaluated using accuracy, precision, recall, F1 score, and the area under the receiver operating characteristic curve (ROC AUC).ResultsThe results indicated that residence location, alcohol consumption, life satisfaction, depressive symptoms, and education level are key factors influencing the risk of MCI among older adults with functional disability. Among the models, the neural network achieved the best overall performance (Accuracy: 0.71, Precision: 0.70, Recall: 0.74, F1 Score: 0.72, ROC AUC: 0.80) with stable results across both the training and test sets.ConclusionThis study provides a scientific tool for the early screening of MCI in older adults with functional disability and offers an efficient and scalable predictive model for clinical applications and community health services.

The accurate measurement of perforation length is important for better fluid flow and cost management. For the past eight decades, research has focused on well perforators within both ballistics science and energy domains. Past years saw numerous standardized testing and empirical model development to estimate perforation penetration depth under downhole conditions. The existing models possess restricted functionality while needing regular calibration and ignore numerous components that influence the penetration depth of perforations. This research aims to create an adaptable machine learning system based on API-19B standard perforator data obtained from different operational environments for perforation penetration length prediction. The holistic machine learning methodology allowed us to create ten machine learning models from normal perforation operational data that includes shot phasing, shot density, casing grade, casing nominal weight, casing outer diameter, explosive type, temperature rating and explosive weight, cement compressive strength, and gun diameter. The created models utilize penetration lengths directly measured through API-19B Section-01 testing, which serve as their output data. The paper implements Gradient Boosting, AdaBoost, Random Forest, Support Vector Machines, Decision Trees, K-Nearest Neighbor, Linear Regression, Neural Network, and Stochastic Gradient Descent algorithms, which received data from 1,648 actual API-19B Section-01 tests. The dataset consists of 16,480 points, which are divided into two sections where 80% (13,184 points) serve training algorithms and 20% (3,296 points) evaluate their predictive capacity. Moreover, the machine learning model's efficiency is evaluated through both K-fold and random sampling validation techniques. The computation of mean absolute percent error (MAPE) revealed the most effective machine learning models, which included AdaBoost, Random Forest, Gradient Boosting, Neural Network (L-BFGS), and K-Nearest Neighbors at 3.3%, 4.5%, 5.3%, and 8.1%, respectively, compared to actual measurements of perforation penetration length. In addition, the models demonstrate high correlation rates (R²) with 0.92, 0.88, 0.86, 0.84, and 0.69, respectively. This paper presents the operational improvements achieved through using machine learning models for estimating perforation penetration length. A machine learning modeling system provides precise, rapid, and economic estimation of perforation penetration length through an easier approach than either API-19B Section-01 tests or empirical models. These machine learning models have the capability to process multiple gun parameters along with different well completion types, which solved a universal problem that empirical models could not manage.

The purpose of this study is to identify the most effective supervised machine learning models for predicting the financial performance of companies listed on the BIST100 index. In the rapidly evolving field of financial forecasting, machine learning techniques offer robust predictive capabilities. This research evaluates a range of supervised models, including Tree-Based Models (Decision Trees, Bagging, Random Forests, Adaboost, Gradient Boosting Machine (GBM), Light-GBM, XGBoost, CatBoost), Neural Network-based Models (Artificial Neural Networks (ANN), Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), Long Short-Term Memory networks (LSTM)), and Instance-based Learning Models (K-Nearest Neighbors (KNN) and Support Vector Machines (SVM)). The models’ performance is assessed using comprehensive error metrics such as Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and relative Root Mean Squared Error (rRMSE). The findings reveal that no single machine learning model consistently outperforms others across all companies in the BIST100 index. However, models like XGBoost and Random Forests demonstrate strong and consistent performance, making them particularly effective for financial performance forecasting. Furthermore, deep learning models such as CNNs, RNNs, and LSTMs show promising results, especially for certain firms. The research highlights key insights for investors and financial analysts seeking to leverage machine learning for data-driven decision-making in the Turkish stock market. This study offers a unique contribution to the field by applying and comparing advanced machine learning techniques in the context of the BIST100 index. It provides actionable insights for improving financial prediction accuracy and offers a foundation for further research in other stock market contexts.

Application of machine learning-based surrogate models for urban flood depth modeling in Ho Chi Minh City, Vietnam

The widespread propagation of wireless communication devices, from smartphones and tablets to Internet of Things (IoT) systems, has become an integral part of modern life. However, the expansion of wireless technology has also raised public concern about the potential health risks associated with prolonged exposure to electromagnetic fields. Our objective is to determine the optimal machine learning model for constructing electric field strength maps across urban areas, enhancing the field of environmental monitoring with the aid of sensor-based data collection. Our machine learning models consist of a novel and comprehensive dataset collected from a network of strategically placed sensors, capturing not only electromagnetic field readings but also additional urban features, including population density, levels of urbanization, and specific building characteristics. This sensor-driven approach, coupled with explainable AI, enables us to identify key factors influencing electromagnetic exposure more accurately. The integration of IoT sensor data with machine learning opens the potential for creating highly detailed and dynamic electromagnetic pollution maps. These maps are not merely static snapshots; they offer researchers the ability to track trends over time, assess the effectiveness of mitigation efforts, and gain a deeper understanding of electromagnetic field distribution in urban environments. Through the extensive dataset, our models can yield highly accurate and dynamic electric field strength maps. For this study, we performed a comprehensive analysis involving 566 machine learning models across eight French cities: Lyon, Saint-Étienne, Clermont-Ferrand, Dijon, Nantes, Rouen, Lille, and Paris. The analysis incorporated six core approaches: k-Nearest Neighbors, XGBoost, Random Forest, Neural Networks, Decision Trees, and Linear Regression. The findings underscore the superior predictive capabilities of ensemble methods such as Random Forests and XGBoost, which outperform individual models. Simpler approaches like Decision Trees and k-NN offer effective yet slightly less precise alternatives. Neural Networks, despite their complexity, highlight the potential for further refinement in this application. In addition, our results show that the machine learning models significantly outperform the linear regression baseline, demonstrating the added value of more complex techniques in this domain. Our SHAP analysis reveals that the feature importance rankings in tree-based machine learning models differ significantly from those in k-NN, neural network, and linear regression models.

Controlling noise pollution in smart cities is a big challenge nowadays due to rise in urbanization and industrialization. As population mass grows, the celebration of yearly festivals such as Dussehra in Bhubaneswar city is also getting popular. However, since this sound pollution is creating a risk to human health, regular monitoring is strictly needed. In this work, the noise pollution level of Bhubaneswar smart city during Dussehra 2020 is predicted using different supervised machine learning (ML) prediction models. The input parameters considered for this work are area or zones of Bhubaneswar city, time at which sound level recorded, equivalent continuous sound level (Leq in dBA), and noise level (high/low compared to the standard value). The data collected for training phase and testing phase by using different ML models is taken from State Pollution Control Board, Odisha, India, for the years 2015–2020. The supervised ML models taken in this work are Decision Tree (DT), Neural Network (NN), k-Nearest Neighbor (k-NN), Naïve Bayes (NB), Support Vector Machine (SVM), and Random Forest (RF). The predictions of the models are evaluated using Orange 3.26 data analytics tool. From the results, it was found that DT and RF show a higher classification accuracy, 92.5%, than that of other ML models. Moreover, it is observed that the probability of prediction of noise pollution level for the testing dataset for DT is higher for high noise level and for RF is higher for low noise level than other prediction models.