Machine Learning Regression Research Articles

Integrating deep learning and regression models for accurate prediction of groundwater fluoride contamination in old city in Bitlis province, Eastern Anatolia Region, Türkiye

Groundwater resources in Bitlis province and its surroundings in Türkiye’s Eastern Anatolia Region are pivotal for drinking water, yet they face a significant threat from fluoride contamination, compounded by the region’s volcanic rock structure. To address this concern, fluoride levels were meticulously measured at 30 points in June 2019 dry period and September 2019 rainy period. Despite the accuracy of present measurement techniques, their time-consuming nature renders them economically unviable. Therefore, this study aims to assess the distribution of probable geogenic contamination of groundwater and develop a robust prediction model by analyzing the relationship between predictive variables and target contaminants. In this pursuit, various machine learning techniques and regression models, including Linear Regression, Random Forest, Decision Tree, K-Neighbors, and XGBoost, as well as deep learning models such as ANN, DNN, CNN, and LSTM, were employed. Elements such as aluminum (Al), boron (B), cadmium (Cd), cobalt (Co), chromium (Cr), copper (Cu), iron (Fe), manganese (Mn), nickel (Ni), phosphorus (Pb), lead (Pb), and zinc (Zn) were utilized as features to predict fluoride levels. The SelectKbest feature selection method was used to improve the accuracy of the prediction model. This method identifies important features in the dataset for different values of k and increases model efficiency. The models were able to produce more accurate predictions by selecting the most important variables. The findings highlight the superior performance of the XGBoost regressor and CNN in predicting groundwater quality, with XGBoost consistently outperforming other models, exhibiting the lowest values for evaluation metrics like mean squared error (MSE), mean absolute error (MAE), and root mean squared error (RMSE) across different k values. For instance, when considering all features, XGBoost attained an MSE of 0.07, an MAE of 0.22, an RMSE of 0.27, a MAPE of 9.25%, and an NSE of 0.75. Conversely, the Decision Tree regressor consistently displayed inferior performance, with its maximum MSE reaching 0.11 (k = 5) and maximum RMSE of 0.33 (k = 5). Furthermore, feature selection analysis revealed the consistent significance of boron (B) and cadmium (Cd) across all datasets, underscoring their pivotal roles in groundwater contamination. Notably, in the machine learning framework evaluation, the XGBoost regressor excelled in modeling both the “all” and “rainy season” datasets, while the convolutional neural network (CNN) outperformed in the “dry season” dataset. This study emphasizes the potential of XGBoost regressor and CNN for accurate groundwater quality prediction and recommends their utilization, while acknowledging the limitations of the Decision Tree Regressor.

Development of machine learning regression models for the prediction of tensile strength of friction stir processed AA8090/SiC surface composites

Friction Stir Processing is a state-of-the-art technology for microstructure refinement, material property enhancement, and fabrication of surface composites. Machine learning approaches have garnered significant interest as prospective models for modeling various production systems. The present work aims to develop four machine learning models, namely linear regression, support vector regression, artificial neural network and extreme gradient boosting to predict the influence of FSP parameters such as tool rotational speed, tool traverse speed and groove width on ultimate tensile strength of friction stir processed AA8090/SiC surface composites. These models were developed through Python programming and the original dataset was divided into 80% for the training phase and 20% for the testing phase. The performance of the models was evaluated by root mean squared error, mean absolute error and R2. Based on the results and graphical visualization, it was observed that the XGBoost model outperformed other models with high accuracy in predicting UTS of AA8090/SiC surface composites.

Estimating Leaf Chlorophyll Content of Winter Wheat from UAV Multispectral Images Using Machine Learning Algorithms under Different Species, Growth Stages, and Nitrogen Stress Conditions

The rapid and accurate estimation of leaf chlorophyll content (LCC), an important indicator of crop photosynthetic capacity and nutritional status, is of great significance for precise nitrogen fertilization management. To explore the existence of a versatile regression model that can be successfully used to estimate the LCC for different varieties under different growth stages and nitrogen stress conditions, a study was conducted in 2023 across the growing season for winter wheat with five species and five nitrogen application levels. Two machine learning regression algorithms, support vector machine (SVM) and random forest (RF), were used to establish the bridge between UAV-derived multispectral vegetation indices and ground truth LCC (relative chlorophyll content, SPAD), taking the multivariate linear regression (MLR) algorithm as a reference. The results show that the visible atmospherically resistant index, vegetative index, and normalized difference vegetation index had the highest correlation with ground truth LCC, with a Pearson’s correlation coefficient of 0.95. All three regression algorithms (MLR, RF, and SVM) performed well on the training dataset (R2: 0.932–0.944, RMSE: 3.96 to 4.37), but performed differently on validation datasets with different growth stages, species, and nitrogen application levels. Compared to winter wheat species and nitrogen application levels, the growth stages had the greatest influence on the generalization ability of LCC estimation models, especially for the dough stage. At the dough stage, compared to MLR and RF, SVM performed best, with R2 increasing by 0.27 and 0.10, respectively, and RMSE decreasing by 1.13 and 0.46, respectively. Overall, this study demonstrated that the combination of UAV-derived multispectral VIs and the SVM regression algorithm could be successfully applied to map the LCC of winter wheat for different species, growth stages, and nitrogen stress conditions. Ultimately, this research is significant as it shows the successful application of UAV data for mapping the LCC of winter wheat across diverse conditions, offering valuable insights for precision nitrogen fertilization management.

Satellite‐Derived Bathymetry in Dynamic Coastal Geomorphological Environments Through Machine Learning Algorithms

AbstractIn the field of coastal geomorphology, advancements in space technology have revolutionized coastal mapping and understanding. Satellite‐derived bathymetry (SDB) research has progressed, focusing on various estimation methods using high‐resolution satellite imagery and in‐situ data. Challenges arise when applying these methods to the Indian coastline due to its turbid waters and intricate features such as creeks and deltas, laden with sediment and submerged rocks. A study aims to assess multivariate machine learning (ML) regression techniques for estimating bathymetric data. The study employs ground‐truth data and imagery from Aster, Landsat‐8, and Sentinel‐2 at distinct sites known for complex underwater landscapes. Several algorithms including Multiple Linear Regression, Support Vector Regressor, Gaussian Process Regression (GPR), Decision Tree Regression, K‐Neighbors Regressor, k‐fold cross‐validation with Decision Tree Regression, and Random Forest (RF) are evaluated for SDB. Results from the Vengurla Site show that using the Landsat‐8 data set with the GPR algorithm achieves R2 0.94, root mean squared error (RMSE) 1.53 m, and MAE 1.14 m, utilizing visible spectrum bands. At Mormugao, the Sentinel‐2 data set with GPR and RF algorithms attains R2 0.97 and RMSE 1.23 m, with GPR outperforming RF, having an MAE of 1.05 m compared to RF's 1.22 m. This study underscores the potential of ML regression techniques in overcoming challenges with using SDB for mapping coastal geomorphology, particularly in intricate underwater terrains and turbid waters by assimilating sophisticated algorithms and their refined cartographic representation.

Assessing Maize Yield Spatiotemporal Variability Using Unmanned Aerial Vehicles and Machine Learning

Optimizing the prediction of maize (Zea mays L.) yields in smallholder farming systems enhances crop management and thus contributes to reducing hunger and achieving one of the Sustainable Development Goals (SDG 2—zero hunger). This research investigated the capability of unmanned aerial vehicle (UAV)-derived data and machine learning algorithms to estimate maize yield and evaluate its spatiotemporal variability through the phenological cycle of the crop in Bronkhorstspruit, South Africa, where UAV data collection took over four dates (pre-flowering, flowering, grain filling, and maturity). The five spectral bands (red, green, blue, near-infrared, and red-edge) of the UAV data, vegetation indices, and grey-level co-occurrence matrix textural features were computed from the bands. Feature selection relied on the correlation between these features and the measured maize yield to estimate maize yield at each growth period. Crop yield prediction was then conducted using our machine learning (ML) regression models, including Random Forest, Gradient Boosting (GradBoost), Categorical Boosting, and Extreme Gradient Boosting. The GradBoost regression showed the best overall model accuracy with R2 ranging from 0.05 to 0.67 and root mean square error from 1.93 to 2.9 t/ha. The yield variability across the growing season indicated that overall higher yield values were predicted in the grain-filling and mature growth stages for both maize fields. An analysis of variance using Welch’s test indicated statistically significant differences in maize yields from the pre-flowering to mature growing stages of the crop (p-value < 0.01). These findings show the utility of UAV data and advanced modelling in detecting yield variations across space and time within smallholder farming environments. Assessing the spatiotemporal variability of maize yields in such environments accurately and timely improves decision-making, essential for ensuring sustainable crop production.

Comparative analysis on the prediction of leak on gas pipeline using physical models and machine learning regression models

Comparative analysis on the prediction of leak on gas pipeline using physical models and machine learning regression models

Evaluating the prognostic value of tumor deposits in non-metastatic lymph node-positive colon adenocarcinoma using Cox regression and machine learning

BackgroundThe 8th AJCC TNM staging for non-metastatic lymph node-positive colon adenocarcinoma patients(NMLP-CA) stages solely by lymph node status, irrespective of the positivity of tumor deposits (TD). This study uses machine learning and Cox regression to predict the prognostic value of tumor deposits in NMLP-CA.MethodsPatient data from the SEER registry (2010–2019) was used to develop CSS nomograms based on prognostic factors identified via multivariate Cox regression. Model performance was evaluated by c-index, dynamic calibration, and Schmid score. Shapley additive explanations (SHAP) were used to explain the selected models.ResultsThe study included 16,548 NMLP-CA patients, randomized 7:3 into training (n = 11,584) and test (n = 4964) sets. Multivariate Cox analysis identified TD, age, marital status, primary site, grade, pT stage, and pN stage as prognostic for cancer-specific survival (CSS). In the test set, the gradient boosting machine (GBM) model achieved the best C-index (0.733) for CSS prediction, while the Cox model and GAMBoost model optimized dynamic calibration(6.473) and Schmid score (0.285), respectively. TD ranked among the top 3 most important features in the models, with increasing predictive significance over time.ConclusionsPositive tumor deposit status confers worse prognosis in NMLP-CA patients. Tumor deposits may confer higher TNM staging. Furthermore, TD could play a more significant role in the staging system.

Quantitative assessment of impact damage in stitched foam‐filled Aluminium honeycomb Sandwich panels by experimental and machine learning methods

AbstractNovel Stitched Foam‐filled Honeycomb Sandwich (SFHS) panels have been fabricated using vacuum‐assisted resin transfer molding to address the weak interfaces between the face sheets and the core in the Foam‐filled Honeycomb Sandwich (FHS) panel. The SFHS panels have shown better load‐bearing capacity and performance characteristics compared to FHS panel after Low‐Velocity Impact (LVI) tests. After the LVI test, MATLAB image processing was used to analyze the impact damage areas and failure mechanisms. In addition, Machine Learning regression algorithms were employed to predict the optimal amount of energy absorbed during low‐velocity impact testing of fabricated panels with a maximum impactor drop height of 700 mm. The results indicated that nylon yarn stitching significantly improved energy absorption and interfacial behavior compared to unstitched honeycomb panels. This research also revealed that SFHS1 panels with adjacent honeycomb cell stitching are more impact resistant, provide increased load carrying capacity, and are cost‐effective. These panels can be utilized by modern engineers to increase economy, durability, and functionality in industrial, automotive, and construction applications.Highlights Stitched Foam Filled Honeycomb Sandwich (SFHS) panels, manufactured via resin transfer molding, and eliminates weak interfaces between the face sheets and core. SFHS panels outperformed unstitched panels in load‐bearing and energy absorption during low‐velocity impact, as confirmed by MATLAB image analysis showing reduced damage and failure. Machine learning algorithms particularly polynomial regression model predicted maximum absorption energy precisely with 99.9% accuracy, close to experimental results. SFHS panels can be used in automotive and industrial applications due to their through‐thickness stitching.

Preserving Sculptural Heritage in the Era of Digital Transformation: Methods and Challenges of 3D Art Assessment

Sustainable digital cultural heritage is now an essential aspect of our lives. The rapid development of 3D technology in the historic preservation industry provides the means of documenting, recovering, and presenting cultural heritage items. However, the digital transformation of 3D sculpture heritage is often led by technology without effective evaluation indicators as a guide. This study compares effective assessment methods for digital forms with traditional art. Our approach uses semantic differential scales and machine learning regression models to assess the importance of fifteen artistic attributes. The semantic differential scale is improved based on 15 artistic attributes and proves to be effective in evaluating the value of digital artwork. This research finds that digital artwork is significantly more popular among young people compared with elderly people, especially for attributes like colour variation, saturation, and texture. The research also finds that complexity and social attributes are more important in predicting the value of the digital 3D model. Digital transformation is a viable method for preserving the artistic value of sculpture and improving cultural sustainability.

Decadal Monitoring of Upwelling Dynamics in Satonda Island Waters Using Landsat-8 and Machine Learning Regression

Global warming and associated weather changes, notably the El Niño Southern Oscillation (ENSO), significantly impact marine ecosystems by altering water quality parameters such as chlorophyll-a (Chl-a) and sea surface temperature (SST). These changes are crucial in understanding the biogeochemical and ecological dynamics of marine environments, especially in regions affected by upwelling. This study aims to monitor upwelling events on Satonda Island, a volcanic island with unique central lake and status as a protected area using remote sensing. Utilizing Landsat-8 imagery and machine learning regression techniques—Random Forest (RF), Support Vector Machine (SVM), and Classification and Regression Tree (CART)—this research evaluates the water quality in Satonda waters over a decade (2013– 2023). The RF method emerged as the most accurate in estimating Chl-a and SST, indicating its efficacy in monitoring marine ecosystems with the result (RMSE = 0.309 and 0.274). The analysis reveals seasonal upwelling patterns, characterized by decreased SST and increased Chl-a concentration, with peaks varying annually between June and November. This study highlights the crucial role of remote sensing and machine learning in monitoring the effects of climate change on marine biodiversity. It provides valuable insights into the temporal dynamics of upwelling in the shallow waters of Indonesia.

Comparison of Synthetic Data Generation Techniques for Control Group Survival Data in Oncology Clinical Trials: Simulation Study.

Synthetic patient data (SPD) generation for survival analysis in oncology trials holds significant potential for accelerating clinical development. Various machine learning methods, including classification and regression trees (CART), random forest (RF), Bayesian network (BN), and conditional tabular generative adversarial network (CTGAN), have been used for this purpose, but their performance in reflecting actual patient survival data remains under investigation. The aim of this study was to determine the most suitable SPD generation method for oncology trials, specifically focusing on both progression-free survival (PFS) and overall survival (OS), which are the primary evaluation end points in oncology trials. To achieve this goal, we conducted a comparative simulation of 4 generation methods, including CART, RF, BN, and the CTGAN, and the performance of each method was evaluated. Using multiple clinical trial data sets, 1000 data sets were generated by using each method for each clinical trial data set and evaluated as follows: (1) median survival time (MST) of PFS and OS; (2) hazard ratio distance (HRD), which indicates the similarity between the actual survival function and a synthetic survival function; and (3) visual analysis of Kaplan-Meier (KM) plots. Each method's ability to mimic the statistical properties of real patient data was evaluated from these multiple angles. In most simulation cases, CART demonstrated the high percentages of MSTs for synthetic data falling within the 95% CI range of the MST of the actual data. These percentages ranged from 88.8% to 98.0% for PFS and from 60.8% to 96.1% for OS. In the evaluation of HRD, CART revealed that HRD values were concentrated at approximately 0.9. Conversely, for the other methods, no consistent trend was observed for either PFS or OS. CART demonstrated better similarity than RF, in that CART caused overfitting and RF (a kind of ensemble learning approach) prevented it. In SPD generation, the statistical properties close to the actual data should be the focus, not a well-generalized prediction model. Both the BN and CTGAN methods cannot accurately reflect the statistical properties of the actual data because small data sets are not suitable. As a method for generating SPD for survival data from small data sets, such as clinical trial data, CART demonstrated to be the most effective method compared to RF, BN, and CTGAN. Additionally, it is possible to improve CART-based generation methods by incorporating feature engineering and other methods in future work.

Quantification of losses in agriculture production in eastern Ukraine due to the Russia-Ukraine war

The ongoing war in Ukraine has seriously impacted the agricultural sector, yet its exact effects on agricultural production are not well understood. Here we combine satellite imagery, machine learning, and statistical regression approaches to present a spatially detailed assessment of agricultural losses for five high-risk provinces in eastern Ukraine (Crimea, Donets’k, Kherson, Luhans’k, and Zaporizhzhya). Our findings indicated that approximately 18.11 ± 2.47% of croplands were left unplanted following the war. Among the cultivated areas, wheat, sunflower, and rapeseed experienced average production losses of 36.39–37.19% in 2022 compared to pre-war levels during 2019–2021. Economically, the indirect losses resulting from decreased production, corresponding to $520.36 ± 22.52, $427.59 ± 24.62, and $205.02 ± 11.53 million USD for wheat, sunflower, and rapeseed crops, respectively, were 1.31–2.16 times higher than direct losses due to unplanted. We also found that reductions in crop production were primarily affected by war intensity indicated by changes in nighttime lights. This methodology offers a comprehensive framework for quantifying agricultural damages from wars, which can be applicable beyond the Russia-Ukraine context.

Treatment of highly iron-contaminated brackish groundwater by MF, UF, and NF membranes for potable water resource production: influence of operating parameters, comparative study, membrane fouling, and machine learning modeling

ABSTRACT The present investigation focuses on using commercially available MF, UF, and NF membranes to eliminate Fe from real groundwater. The impact of process parameters, including applied pressure, temperature, pH, time, and concentration, on flux and Fe removal% is investigated. Results of the permeation test confirm higher permeability for MF membranes (214.71 L/m2.h.bar) than that for NF (2.708 L/m2.h.bar) and UF (56.52 L/m2.h.bar) membranes. The FESEM-EDS characterization confirms the deposition of dominant foulant Fe particles over the membrane surface. At 2 bar applied pressure (temperature = 30 °C, pH = 6.13, and concentration = 6.62 ppm), MF, UF, and NF can eliminate 93.5, 90.2, and 100% Fe, respectively. However, UF and NF exhibit much lower fluxes (72.43 and 3.85 L/m2.h, respectively) compared to MF (317.13 L/m2.h) at the same process conditions. NF eliminates 100% Fe from start to end with rising pressure, while MF (62.16–100) and UF (87.09–100) show gradually improved removal rates. Above pH 6, all membranes demonstrate higher Fe rejection because the oxidation rate of ferrous iron increases at a high pH. With increasing concentration (6–500 ppm), Fe removal efficiency increases for NF (88.26–94.18%) but decreases for both MF (89.63–23.64%) and UF (27.66–14.4%). To validate experimental findings, five machine learning (ML) regression models are scrutinized using various statistical measures.

Development of automatic tuning for combined preprocessing and hyperparameters of machine learning and its application to NIR spectral data of coconut milk adulteration

This study proposed a novel approach to automatically select the preprocessing methods and hyperparameters of machine learning (ML) algorithms based on their best performance in cross-validation for near-infrared (NIR) spectroscopy data. The proposed method simultaneously incorporates single or multiple-preprocessing steps and tunes hyperparameters to determine the best model performance for FT-NIR and Micro-NIR spectral data of coconut milk adulteration with distilled water and mature coconut water in the range of 0%–50%. Computational experiments were conducted using nine single preprocessing types, three types of ML classifier (linear discriminant analysis (LDA), k-nearest neighbour (KNN), multilayer perceptron (MLP)) and three types of ML regressor (partial least squares (PLS), KNN, MLP). The proposed performance strategy effectively addressed and produced satisfactory outcomes for classification and regression challenges in coconut milk adulteration. Finally, the results demonstrated that the proposed approach can more accurately determine the best model, particularly for NIR spectroscopy of coconut milk adulteration.

Spatio-Temporal Dynamics Coupling between Land Use/Cover Change and Water Quality in Dongjiang Lake Watershed Using Satellite Remote Sensing

With rapid social and economic development, land use/land cover change (LUCC) has intensified with serious impacts on water quality in the watershed. In this study, we took Dongjiang Lake watershed as the study area and obtained measured data on water quality parameters from the watershed’s water quality monitoring stations. Based on Landsat-5, Landsat-8, or Sentinel-2 remote sensing data for multiple periods per year between 1992 and 2022, the sensitive satellite bands or band combinations of each water quality parameter were determined. The Random Forest method was used to classify the land use types in the watershed into six categories, and the area proportion of each type was calculated. We established machine learning regression models and polynomial regression models with WQI as the dependent variable and the area proportion of each land use type as the independent variable. Accuracy test results showed that, among them, the quadratic cubic polynomial regression model with grassland, forest land, construction land, and unused land as its independent variables was the best model for coupling watershed water quality with LUCC. This study’s results provide a scientific basis for monitoring spatial and temporal changes in water quality caused by LUCC in the Dongjiang Lake watershed.

Comparative analysis of machine learning techniques for metamaterial absorber performance in terahertz applications

Metamaterial absorbers (MMAs) have tremendous potential for controlling and modulating Terahertz electromagnetic (EM) waves. It is challenging to design MMAs for optimal performance using conventional methods due to their time-consuming and computationally demanding nature. Machine learning (ML) regressor approaches have emerged as potential tools for optimizing the predictive modeling of MMAs in recent years. This article examined different regression techniques, such as Decision Tree, k-Nearest Neighbors, Random Forest, Extra Trees (ET), Extreme Gradient Boosting, Bagging, and Categorical Boosting. The primary goal is to assess the effectiveness of each regressor technique in forecasting the performance of MMAs using various performance metrics. The study focuses on a representative spectrum of MMAs, chosen for its key features that are common across a wide range of MMAs. The extra tree Regressor outperforms other models in the comparative analysis due to its capacity to integrate multiple decision trees and include randomness into the feature selection process, resulting in improved predictive skills. The ET regressor forecasts MMA performance remarkably well for a 70 % test size, with a Root Mean Squared Error of 0.0251, a Mean Absolute Error of 0.0074, a Mean Squared Error of 0.006, and an Adjusted R-squared of 0.9873. These findings can assist researchers in optimizing metamaterial absorber designs for Terahertz applications using ML regressor techniques and provide insights into how these methods can be generalized to other spectra with similar characteristics.

An Architecture as an Alternative to Gradient Boosted Decision Trees for Multiple Machine Learning Tasks

Deep networks-based models have achieved excellent performances in various applications for extracting discriminative feature representations by convolutional neural networks (CNN) or recurrent neural networks (RNN). However, CNN or RNN may not work when handling data without temporal/spatial structures. Therefore, finding a new technique to extract features instead of CNN or RNN is a necessity. Gradient Boosted Decision Trees (GBDT) can select the features with the largest information gain when building trees. In this paper, we propose an architecture based on the ensemble of decision trees and neural network (NN) for multiple machine learning tasks, e.g., classification, regression, and ranking. It can be regarded as an extension of the widely used deep-networks-based model, in which we use GBDT instead of CNN or RNN. This architecture consists of two main parts: (1) the decision forest layers, which focus on learning features from the input data, (2) the fully connected layers, which focus on distilling knowledge from the decision forest layers. Powered by these two parts, the proposed model could handle data without temporal/spatial structures. This model can be efficiently trained by stochastic gradient descent via back-propagation. The empirical evaluation results of different machine learning tasks demonstrate the the effectiveness of the proposed method.

Predicting Oil Production After Enhancement Techniques Using Multidimensional Feature Representation Learning: A Case Study of Profile Control Technique

Summary The prediction of oil production following enhancement techniques has garnered widespread attention, leading scientists to explore this area using machine learning. However, field data collection constraints and single model accuracy limitations mean few models can precisely predict daily oil production after technique implementation. Building upon previous research, this paper introduces a model that predicts oil production after enhancement operations, utilizing multidimensional feature representation learning. It thoroughly examines three characteristic categories affecting the effectiveness of oil production enhancement techniques: geological static parameters, production dynamic parameters, and enhancement technique process parameters. The model comprehensively explores these features with an emphasis on global spatial, local spatial, and temporal information. A complete machine learning prediction process is established, which includes data preprocessing, model training, cross-validation, and oil production prediction after implementing enhancement techniques. The first part of the model involves representation learning on processed data, producing three sets of new features: global spatial, local spatial, and temporal information. These features are fused with the original data, serving as input for the advanced ensemble learning model XGBoost, which predicts daily oil production after implementing the technique. Following the construction of the model, actual field data from profile control techniques are selected to conduct various evaluations based on the model’s performance on validation and test sets. Compared with traditional machine learning regression algorithms, this model demonstrates significantly higher predictive accuracy. The prediction accuracy for oil production using given enhanced techniques reached 96% in the validation set and 94% in the test set. This research provides a technical foundation for selecting appropriate production enhancement techniques in oil fields by accurately predicting oil production after implementing enhancement techniques, which offers guidance for actual oilfield production.

Comparison of Machine Learning Algorithms for Predicting Stunting Prevalence in Indonesia

Stunting is a serious public health problem, especially among under-fives, which can cause serious short- and long-term impacts. Efforts to tackle stunting in Indonesia involve national strategies and development priorities. Therefore, this study aims to compare the performance of machine learning regression algorithms in predicting stunting prevalence in Indonesia. The data collected is secondary data. The data collection was done carefully, taking explicit details regarding the source, scope, extent, and analysis of the dataset, and using a careful sampling methodology. The model evaluation results show that the Random Forest Regression algorithm has the best performance, with a success rate of 90.537%. The application of this model to the new dataset shows that East Nusa Tenggara province has the highest percentage of stunting at 31.85%, while Bali has the lowest percentage at 12.07%. Visualization of the dashboard using Tableau provides a clear picture of the distribution of stunting in Indonesia. In conclusion, this research contributes to the development of science, especially in the field of machine learning and public health, and provides policy recommendations for tackling stunting in Indonesia.

Dynamic interactions between oil prices and renewable energy production in Italy amid the COVID-19 pandemic: wavelet and machine learning analyses

This study examines the intricate dynamics between oil prices and renewable energy investments in Italy during the initial phase of the CoronaVirus Disease 2019 pandemic, a period characterized by significant economic and social upheaval. Utilizing advanced empirical techniques, such as Partial Wavelet Coherency analysis, Time-Varying Granger Causality, and Robinson Log-Periodogram tests, as well as Machine Learning (ML) regressions, this research uncovers nuanced insights into the interplay between oil prices and renewable energy series including biomass, solar, hydro, wind, and geothermal. Key findings indicate a predominant in-phase relationship with oil prices leading most renewable energy series, and unidirectional causality from renewables to oil prices in several instances, highlighting the potential influence of renewable energy on oil market dynamics. In robustness checks, ML models further elucidate the impact, with solar, hydro, and geothermal sources showing significant importance scores. These insights are critical for policymakers and stakeholders aiming to enhance energy security and transition towards sustainable energy sources amidst global crises.