Forest Approach Research Articles

Forecasting crude oil price: A deep forest ensemble approach

We made an application of a cutting-edge machine learning method, the deep forest ensemble approach (DFEA, Zhou and Feng (2017, 2019)), to empirically predict crude oil prices. We used a large data set with 36 explanatory variables to compare the predictability of the DFEA with seven popular machine learning models and their mean combination method. The out-of-sample forecasting results showed that the DFEA statistically and economically outperforms all the competing models in terms of out-of-sample R square and success ratio. The DFEA also displayed sizable certainty equivalent return (CER) gains for a mean-variance investor in practice from an asset allocation perspective. Furthermore, we found that the predictive power of the DFEA stems from technical indicators, especially momentum predictors. Our results survived in various robustness checks.

Impact of serum potassium on length of hospitalization stay: A random forest approach

Impact of serum potassium on length of hospitalization stay: <scp>A</scp> random forest approach

Analysis of Road Roughness and Driver Comfort in 'Long-Haul' Road Transportation Using Random Forest Approach.

Global trade depends on long-haul transportation, yet comfort for drivers on lengthy trips is sometimes neglected. Rough roads have a major negative influence on driver comfort and increase the risk of weariness, distracted driving, and accidents. Using Random Forest regression, a machine learning technique well-suited to examining big datasets and nonlinear relationships, this study examines the relationship between road roughness and driver comfort. Using the MIRANDA mobile application, data were gathered from 1,048,576 rows, including vehicle acceleration and values for the International Roughness Index (IRI). The Support Vector Regression (SVR) and XGBoost models were used for comparative analysis. Random Forest was preferred because of its ability to be deployed in real time and use less memory, even if XGBoost performed better in terms of training time and prediction accuracy. The findings showed a significant relationship between driver discomfort and road roughness, with rougher roads resulting in increased vertical acceleration and lower comfort levels (Road Roughness: SD-0.73; Driver's Comfort: Mean-10.01, SD-0.64). This study highlights how crucial it is to provide smooth surfaces and road maintenance in order to increase road safety, lessen driver weariness, and promote long-haul driver welfare. These results offer information to transportation authorities and policymakers to help them make data-driven decisions that enhance the efficiency of transportation and road conditions.

Data-Driven Strategies for Carbimazole Titration: Exploring Machine Learning Solutions in Hyperthyroidism Control.

University Hospitals Dorset (UHD) has over 1,000 thyroid patient contacts annually. These are primarily patients with autoimmune hyperthyroidism treated with Carbimazole titration. Dose adjustments are made by a healthcare professional (HCP) based on the results of thyroid function tests, who then prescribes a dose and communicates this to the patient via letter. This is time-consuming and introduces treatment delays. This study aimed to replace some time-intensive manual dose adjustments with a machine learning model to determine Carbimazole dosing. This can in the future serve patients with rapid and safe dose determination and ease the pressures on HCPs. Data from 421 hyperthyroidism patients at UHD were extracted and anonymised. A total of 353 patients (83.85%) were included in the study. Different machine-learning classification algorithms were tested under several data processing regimes. Using an iterative approach, consisting of an initial model selection followed by a feature selection method the performance was improved. Models were evaluated using weighted F1 scores and Brier scores to select the best model with the highest confidence. The best performance is achieved using a random forest (RF) approach, resulting in good average F1 scores of 0.731. A model was selected based on a balanced assessment considering the accuracy of the prediction (F1 = 0.751) and the confidence of the model (Brier score = 0.38). To simulate a use-case, the accumulation of the prediction error over time was assessed. It was determined that an improvement in accuracy is expected if this model was to be deployed in practice.

Electronic Nose Humidity Compensation System Based on Rapid Detection.

In this study, we present an electronic nose (e-nose) humidity compensation system based on rapid detection to solve the issue of humidity drift's potential negative impact on the performance of electronic noses. First, we chose the first ten seconds of non-steady state (rapid detection mode) sensor data as the dataset, rather than waiting for the electronic nose to stabilize during the detection process. This was carried out in the hope of improving the detection efficiency of the e-nose and to demonstrate that the e-nose can collect gasses efficiently in rapid detection mode. The random forest approach is then used to optimize and reduce the dataset's dimensionality, filtering critical features and improving the electronic nose's classification capacity. Finally, this study builds an electronic nose humidity compensation system to compensate for the datasets generated via rapid real-time detection, efficiently correcting the deviation of the sensor response caused by humidity variations. This method enhanced the average resolution of the electronic nose in this trial from 87.7% to 99.3%, a 12.4% improvement, demonstrating the efficacy of the humidity compensation system based on rapid detection for the electronic nose. This strategy not only improves the electronic nose's anti-drift and classification capabilities but also extends its service life, presenting a new solution for the electronic nose in practical detecting applications.

Remote Prediction of Soybean Yield Using UAV-Based Hyperspectral Imaging and Machine Learning Models

Early soybean yield estimation has become a fundamental tool for market policy and food security. Considering a heterogeneous crop, this study investigates the spatial and spectral variability in soybean canopy reflectance to achieve grain yield estimation. Besides allowing crop mapping, remote sensing data also provide spectral evidence that can be used as a priori knowledge to guide sample collection for prediction models. In this context, this study proposes a sampling design method that distributes sample plots based on the spatial and spectral variability in vegetation spectral indices observed in the field. Random forest (RF) and multiple linear regression (MLR) approaches were applied to a set of spectral bands and six vegetation indices to assess their contributions to the soybean yield estimates. Experiments were conducted with a hyperspectral sensor of 25 contiguous spectral bands, ranging from 500 to 900 nm, carried by an unmanned aerial vehicle (UAV) to collect images during the R5 soybean growth stage. The tests showed that spectral indices specially designed from some bands could be adopted instead of using multiple bands with MLR. However, the best result was obtained with RF using spectral bands and the height attribute extracted from the photogrammetric height model. In this case, Pearson’s correlation coefficient was 0.91. The difference between the grain yield productivity estimated with the RF model and the weight collected at harvest was 1.5%, indicating high accuracy for yield prediction.

Long‐term assessment of macro‐ and micronutrients in foliage of European beech (Fagus sylvatica L.) in thinned versus unmanaged old‐growth stands

AbstractBackgroundScience‐based decisions regarding forest management require the knowledge of the impact of thinning regimens on the forests’ vitality and resilience. There is no systematic study analysing the role of forest management approaches on the nutritional status of forests, serving as a surrogate for their health and growth.AimsWe assessed the impact of ‘heavy thinning from above’ versus ‘no management’ on the foliar chemistry of old‐growth European beech stands on a calcareous site with cambisol/chromic luvisol soil in Thuringia, Germany.MethodsMacro‐ and micronutrients were analysed by serial foliar analysis of six trees per experimental plot over 13 years (2009–2021). To assess potential differences of foliar chemistry between the two plots and over time, a linear mixed‐effects model was applied.ResultsFoliar concentrations of all macro‐ and micronutrients were not significantly different between the two plots (p &gt; 0.05), demonstrating that the management approach had no relevant impact on the nutritional status of beech trees growing at the calcareous site. Furthermore, all foliar concentrations were dynamic over the 13‐year evaluation period. Hence, long‐term forest monitoring is crucial to capture the complex interplay between the trees and environmental conditions.ConclusionsSerial foliar analysis allows for a reliable evaluation of a forest's nutritional status. The results indicate that either regimen, that is, ‘heavy thinning from above’ or ‘no management’, shall not pose any risk in terms of growth and stability. Our results add to the understanding of beech forest dynamics and may provide a further piece for science‐based strategies of sustainable forest management.

Does Increasing Concentration Hit Poorer Areas More? A Study of Retail Petroleum Markets*

A central tenet in the field of industrial organisation is that increasing/decreasing market concentration is associated with increased/reduced markups. But does this variation affect every consumer to the same extent? Previous literature finds price dispersion exists even for homogeneous goods, at least partially as a result of heterogeneity in consumer engagement with the market. We study this question by linking demographic and income heterogeneity across local areas to the impact of changing market concentration on markups. With 15 years of station‐level motor fuel price data from Western Australia and information on instances of local market exit and entry, we apply a non‐parametric causal forest approach to explore the heterogeneity in the effect of exit/entry. The paper provides evidence of the distributional effect of changing market concentration. Areas with lower income experience a larger increase in petrol stations' price margin as a result of market exit. On the other hand, entry does not benefit the same low‐income areas with a larger reduction in the margin than in high‐income areas. Policy implications include a need to further focus on increasing engagement by low‐income consumers.

Statistical characteristics of aquitard hydraulic conductivity, specific storage and porosity

Aquitard hydraulic conductivity (K), specific storage (Ss) and porosity (ϕ) are critical to the modeling of flow and mass and heat transport in groundwater systems, yet comprehensive insights into their statistical characteristics remain limited. This study compiles 456 K values from 47 studies, 202 Ss values from 49 studies and 239 ϕ values from 18 studies, followed by rigorous statistical analysis of aquitard K, Ss and ϕ with dependence on the features of test method, lithology and burial depth. Additionally, the interrelationships among K, Ss and ϕ are analyzed. A predictive model for aquitard K and Ss is developed utilizing the random forest (RF) approach. The statistical findings unearth that laboratory experiments tend to yield lower K while higher Ss, particularly when burial depths are less than 50 m. Clay, silt clay, and glacial till exhibit comparable K values, in contrast to significantly lower K values observed in shale formations. Aquitard K, Ss, and ϕ exhibit depth-decaying characteristics. Between two adjacent depth intervals, 0 ∼ 50 m, 50 ∼ 150 m and >150 m, there is nearly an order of magnitude difference in either K or Ss. Aquitard K tends to increase with ϕ when ϕ is below 0.25 and decreases vice versa. This phenomenon is probably attributed to the significant amount of immobile-bound water adhering to clayey particles within the aquitard with a high ϕ value, which reduces the pore space available for water flow. Elastic Ss, which indicates recoverable aquitard groundwater depletion, is about one order of magnitude lower than inelastic Ss, which indicates permanent aquitard groundwater depletion, implying that the majority of groundwater depletion from aquitards under long-term groundwater exploitation is non-recoverable. The RF-based predictive model indicates that ϕ is a highly important feature for predicting K and outperforms lithology, test method and burial depth in predicting Ss. The statistical findings of this study present a valuable basis for aquitard hydraulic parameters.

Enhanced Fault Diagnosis in IoT: Uniting Data Fusion with Deep Multi-Scale Fusion Neural Network

One of the biggest challenges is figuring out when industrial IoT devices break. Notwithstanding these difficulties, one of the many advantages of using real-time sensor data from the Industrial Internet of Things (IIoT) is the ability to monitor events and respond promptly. The IIoT's sensor numbers may be analysed, combined with data from other sources, and applied quickly to make decisions. This manuscript proposes Enhanced Fault Diagnosis in IoT, Uniting Data Fusion with Deep Multi-Scale Fusion Neural Network (FD-IoT-DMSFNN). Initially, input sensor data are taken from the CWRU Dataset. Then, sensor data is normalised using Multivariate Fast Iterative Filtering. Then, the normalised sensor data are extensively dispersed and diverse; hence, the data outlier detection is performed using the Deep isolation forest (DIF) approach. Therefore, this manuscript proposes a Mexican Axolotl Optimization (MAO) for tuning the weight parameter of DMSFNN, which exhibits an enhanced fault detection process. Python is used to implement the recommended strategy. The proposed method's performance yields a lower completion time of 6.5%, 1.8%, and 4.13%; higher prediction accuracy of 2.26%, 6.71%, and 8.32% compared with existing approaches such as towards deep domain adaptability training methods for industrial IoT edge device soft real-time fault diagnosis (FD-IoT-LSTM), Intelligent Fault Quantitative Identification for the Industrial Internet of Things (IIoT) utilising a particular deep dual reinforcement learning model with insufficient samples (FD-IoT-DRL) and IIoT- based fault identification model for industrial use (FD-IoT-ML-LSTM) respectively.

Exploring Heterogeneity in the Cost-Effectiveness of High-Flow Nasal Cannula Therapy in Acutely Ill Children—Insights From the Step-Up First-line Support for Assistance in Breathing in Children Trial Using a Machine Learning Method

ObjectivesTo investigate heterogeneity in the cost-effectiveness of high-flow nasal cannula (HFNC) therapy compared with continuous positive airway pressure (CPAP) for acutely ill children requiring noninvasive respiratory support. MethodsUsing data from the First-line Support for Assistance in Breathing in Children trial, we explore heterogeneity at the patient and subgroup levels using 2 causal forest approaches and a seemingly unrelated regression approach for comparison. First-line Support for Assistance in Breathing in Children is a noninferiority randomized controlled trial (ISRCTN60048867) involving 24 UK pediatric intensive care units. The Step-up trial focuses on acutely ill children aged 0 to 15 years, requiring noninvasive respiratory support. A total of 600 children were randomly assigned to HFNC and CPAP groups in a 1:1 allocation ratio, with 94 patients excluded because of data unavailability. ResultsThe primary outcome is the incremental net monetary benefit (INB) of HFNC compared with CPAP, using a willingness-to-pay threshold of £20 000 per quality-adjusted life year gain. INB is derived from total costs and quality-adjusted life years at 6 months. Subgroup analysis showed that some subgroups, such as male children, those aged less than 12 months, and those without severe respiratory distress at randomization, had more favorable INB results. Patient-level analysis revealed heterogeneity in INB estimates, particularly driven by the cost component, with greater uncertainty for those with higher INBs. ConclusionsThe estimated overall INB of HFNC is significantly larger for specific patient subgroups, suggesting that the cost-effectiveness of HFNC can be heterogeneous, which highlights the importance of considering patient characteristics in evaluating the cost-effectiveness of HFNC.

Performance Comparison of Support Vector Machines, AdaBoost, and Random Forest for Sentiment Text Analysis and Classification

In sentiment analysis, text analysis becomes an important process to derive useful information from the unstructured data. In this work, we study the performance of three advanced machine learning algorithms, Support Vector Machines (SVM), Random Forest, and AdaBoost, for a specific sentiment classification task. Each classifier was trained and evaluated on principal metrics such as Area Under the Curve (AUC), Classification Accuracy (CA), F1-Score, Precision, and Recall using the Bag of Words model for feature extraction. Those results show that the Random Forest approach beat both SVM and AdaBoost, with an AUC of 0.988, CA = 0.915, and F1-Score = 0.915 SVM demonstrated moderate performance with an AUC of 0.939 and an F1-Score of 0.845, while AdaBoost exhibited the worst performance in all metrics based on that ensemble model-based classifiers for data change predictions. Random Forest may thus be a powerful machine learning technique to implement for sentiment analysis in text.

Assessing and mapping of soil organic carbon at multiple depths in the semi-arid Trans-Ural steppe zone

The quantification of soil organic carbon (SOC) and its vertical distribution is crucial for understanding carbon dynamics in terrestrial ecosystems. This study aimed to 2.5D digital mapping of SOC content in the Trans-Ural steppe zone (Russia) using a quantile regression forest (QRF) approach. The study utilized a dataset comprising 2495 SOC measurements collected from 1316 locations across three soil depths: 0–20 cm, 20–40 cm, and 40–60 cm. Environmental covariates were incorporated into the SOC modeling process, capturing major soil formation factors, and the uncertainty of the generated maps was estimated. The results revealed that SOC content ranged from 0.59 to 9.05 % in the topsoil, from 0.5 to 6.61 % in the subsurface layer and from 0.06 to 4.64 % in the subsoil. Based on the error metrics, including root mean square error (RMSE), coefficient of determination (R2), and Nash-Sutcliffe efficiency coefficient (NSE), we found a decrease in prediction accuracy with increasing soil depth. Furthermore, climate and vegetation variables, as well as elevation, emerged as key factors influencing the prediction of SOC concentrations at all depths. We also made an attempt to assess the future change of SOC under the influence of climate and anthropogenic impact. We anticipate that climate aridization and plowing will lead to a decline in SOC content in the Trans-Ural steppe region. Our findings contribute to the existing knowledge of SOC dynamics in steppe ecosystems at several depths, supporting informed decision-making for sustainable land use and climate change mitigation strategies.

SCARF: A new algorithm for continuous prediction of biomass dynamics using machine learning and Landsat time series

We developed the SCARF (Spatial Mismatch and Systematic Prediction Error Corrected cAscade Random Forests) algorithm for continuous prediction of biomass dynamics using machine learning and Landsat Time Series (LTS). Our approach addresses the challenges posed by the cloudy subtropical forests in southern China, where monitoring biomass dynamics is notoriously difficult. To derive spectral-temporal features from the LTS, we applied the Continuous Change Detection and Classification (CCDC) algorithm (Zhu and Woodcock, 2014). Subsequently, we employed the cascade random forests machine learning algorithm for biomass prediction. This new approach corrects the spatial mismatch effects between plots and Landsat pixels as well as the systematic prediction errors in the machine learning model. As a result, it substantially enhances biomass prediction accuracy, with a coefficient of determination (R2) of 0.83 and a root mean square error (RMSE) of 6.27 Mg ha-1. In comparison, the commonly used random forests approach yields an R2 of 0.47 and RMSE of 8.52 Mg ha-1. Additionally, it provides reliable spatial prediction beyond the model-training area, achieving an R2 of 0.79 and an RMSE of 6.62 Mg ha-1. Furthermore, we demonstrate that modeling five different forest age groups separately further improves prediction accuracies, resulting in an increased R2 of 0.87 and a reduced RMSE of 3.65 Mg ha-1. A comparison of the allometric model prediction from the field plots and those from the SCARF model revealed a strong agreement, indicating that this approach can provide a temporally continuous prediction of biomass dynamics. Our study presents a robust method for continuous, reliable, and explicit spatiotemporal prediction of biomass dynamics in cloudy subtropical forests using LTS.

Using machine-learning approaches to investigate the volatile-compound fingerprint of fishy off-flavour from beef with enhanced healthful fatty acids

Machine learning classification approaches were used to discriminate a fishy off-flavour identified in beef with health-enhanced fatty acid profiles. The random forest approach outperformed (P < 0.001; receiver operating characteristic curve: 99.8 %, sensitivity: 99.9 % and specificity: 93.7 %) the logistic regression, partial least-squares discrimination analysis and the support vector machine (linear and radial) approaches, correctly classifying 100 % and 82 % of the fishy and non-fishy meat samples, respectively. The random forest algorithm identified 20 volatile compounds responsible for the discrimination of fishy from non-fishy meat samples. Among those, seven volatile compounds (pentadecane, octadecane, γ-dodecalactone, dodecanal, (E,E)-2,4-heptadienal, 2-heptanone, and ethylbenzene) were selected as significant contributors to the fishy off-flavour fingerprint, all being related to lipid oxidation. This fishy off-flavour fingerprint could facilitate the rapid monitoring of beef with enhanced healthy fatty acids to avoid consumer dissatisfaction due to fishy off-flavour.

Impact of wetland health and provisioning services on the livelihood of the fishing community.

The present study aimed to explore the linkage between wetland health, provisioning service value (PsV) and livelihood vulnerability of the dependent fishermen community taking examples from the Moribund deltaic wetlands of India. Wetland health including hydrological strength, habitat state, and the water quality of the wetland was assessed using a random forest (RF) and XGBoosing machine learning approach, and the Livelihood Vulnerability Index (LVI) was computed using balanced weighted approach. Ordinary least square (OLS) regression, correlation matrix, and rank correlation matrix were done to make the relationship between wetland health and LVI. Wetland health was found poor (28.38%) in the isolated, smaller, and peripheral parts of the wetland with agricultural and built-up area proximity. Hydrological strength (with r2 of 0.67) was found as the most dominant health determinant followed by habitat state (with r2 of 0.62). OLS reported that in most of the cases, the standard regression residual is low (0.5 to - 0.5) which indicates that there is a strong relation between wetland health and LVI. KDE plot and correlation matrix also figured out the same. From the field survey, it was found that the wetlands with good habitat health are promising for providing more provisioning services like fish which in turn supports the livelihood of the dependent communities. The findings of this study have a deeper insight into livelihood management through wetland management. Hence, it would inspire policymakers and stakeholders to conserve wetlands not only for the sake of ecology but also for society.

Using machine learning techniques for exploration and classification of laboratory data

Abstract Objectives The study aims to acquaint readers with six widely used machine learning (ML) techniques (Principal Component Analysis (PCA), Uniform Manifold Approximation and Projection (UMAP), k-means, hierarchical clustering and the decision tree models (rpart and random forest)) that might be useful for the analysis of laboratory data. Methods Utilizing a recently validated data set from lung cancer diagnostics, we investigate how ML can support the search for a suitable tumor marker panel for the differentiation of small cell (SCLC) and non-small cell lung cancer (NSCLC). Results The ML techniques used here effectively helped to gain a quick overview of the data structures and provide initial answers to the clinical questions. Dimensionality reduction techniques such as PCA and UMAP offered insightful visualization and impression of the data structure, suggesting the existence of two tumor groups with a large overlap of largely inconspicuous values. This impression was confirmed by a cluster analysis with the k-means algorithm, indicative of unsupervised learning. For supervised learning, decision tree models like rpart or random forest demonstrated their utility in differential diagnosis of the two tumor types. The rpart model, which constructs binary decision trees based on the recursive partitioning algorithm, suggests a tree involving four serum tumor markers (STMs), which were confirmed by the random forest approach. Both highlighted pro-gastrin-releasing peptide (ProGRP), neuron specific enolase (NSE), cytokeratin-19 fragment (CYFRA 21-1) and cancer antigen (CA) 72-4 as key tumor markers, aligning with the outcomes of the initial statistical analysis. Cross-validation of the two proposals showed a higher area under the receiver operating characteristic (AUROC) curve of 0.95 with a 95 % confidence interval (CI) of 0.92–0.97 for the random forest model compared to an AUROC curve of 0.88 (95 % CI: 0.83–0.93). Conclusions ML can provide a useful overview of inherent medical data structures and distinguish significant from less pertinent features. While by no means replacing human medical and statistical expertise, ML can significantly accelerate the evaluation of medical data, supporting a more informed diagnostic dialogue between physicians and statisticians.

Prediction of heart disease using a novel slap swarm optimized multi-objective random forest

Heart disease is a leading cause of death worldwide, becoming a major health concern for many people. Among the most important tasks and regular diagnostic research entails the identified cardiac issues, like heart diseases, valve conditions, etc. Early detection of heart disease may save many lives. The application of machine learning techniques in the medical sector has advanced significantly. A novel Slap Swarm Optimized Multi-Objective Random Forest (SSO-MORF) approach was presented in the proposed work for predicting cardiac disease. For this suggested investigation, heart disease information using individual customer identification at the University of California Irvine (UCI) was utilized, and data mining methods like categorization were applied. Min–max normalization and principal component analysis (PCA) were used for data preparation and feature extraction. As a result, a rather basic supervised machine learning technique may be used to predict heart disease quite accurately and with excellent potential value. According to this study, a classification based on the RF technique produced results with 99.45% accuracy, 98.3% recall, 98.6% precision, and a 98.9% f1 score. This finding demonstrates that our system is more efficient at predicting cardiac disease than other cutting-edge techniques.

Monte Carlo Simulation, Artificial Intelligence and Machine Learning-based Modelling and Optimization of Three-dimensional Electrochemical Treatment of Xenobiotic Dye Wastewater

The present study investigates the synergistic performance of the three-dimensional electrochemical process to decolourise methyl orange (MO) dye pollutant from xenobiotic textile wastewater. The textile dye was treated using electrochemical technique with strong oxidizing potential, and additional adsorption technology was employed to effectively remove dye pollutants from wastewater. Approximately 98% of MO removal efficiency was achieved using 15 mA/cm2 of current density, 3.62 kWh/kg of energy consumption and 79.53% of current efficiency. The 50 mg/L MO pollutant was rapidly mineralized with a half-life of 4.66 min at a current density of 15 mA/cm2. Additionally, graphite intercalation compound (GIC) was electrically polarized in the three-dimensional electrochemical reactor to enhance the direct electrooxidation and.OH generation, thereby improving synergistic treatment efficiency. Decolourisation of MO-polluted wastewater was optimized by artificial intelligence (AI) and machine learning (ML) techniques such as Artificial Neural Networks (ANN), Support Vector Machine (SVM), and random forest (RF) algorithms. Statistical metrics indicated the superiority of the model followed this order: ANN > RF > SVM > Multiple regression. The optimization results of the process parameters by artificial neural network (ANN) and random forest (RF) approaches showed that a current density of 15 mA/cm2, electrolysis time of 30 min and initial MO concentration of 50 mg/L were the best operating parameters to maintain current and energy efficiencies of the electrochemical reactor. Finally, Monte Carlo simulations and sensitivity analysis showed that ANN yielded the best prediction efficiency with the lowest uncertainty and variability level, whereas the predictive outcome of random forest was slightly better.Highlights• In-depth analysis of various artificial intelligence optimization techniques.• Prediction efficiency of artificial intelligence and machine learning algorithms.• 98% dye removal and 100% regeneration of graphite intercalation compound.• Advanced statistical analysis of targeted responses and data fitting techniques.• Analysis of uncertainties and variability using Monte Carlo simulation.

Identification of response regulation governing ozone formation based on influential factors using a random forest approach

The pursuit of enhanced scientific, refined, and precise ozone and air quality control continues to pose significant challenges. Using data visualization techniques and random forest (RF) algorithms, the temporal distribution of atmospheric pollutants and the interrelationship between O3 concentration and its influential factors were investigated with one-year monitoring data in Deqing county in 2021. The local atmospheric conditions predominantly belonged to NOx-sensitive and transition zone. Extremely high O3 concentration were primarily observed when temperatures (T) exceeded 30 °C, with relative humidity (RH) ranging between 30 and 60 %. NO2, RH and T were identified as the top 3 important factors, and O3 concentration have stronger linearly relationship to RH and T, while stronger nonlinearly relationship to NO2. By employing an optimized RF model, controlling consistent mild and high reaction atmospheric conditions, the O3 concentration response to the change of individual influencing factors was acquired. The O3 concentration increased and then decreased in response to the increasing NO2 concentration, displaying a characteristic inflection point at 10 μg m−3. More reactive radicals produced at higher VOCs concentration and continuing NOx cycle at lower NO2 concentration, resulting in the acceleration in the direction of producing more O3. Therefore, the significant different O3 response to variation of VOCs and NOx concentration between mild and high reaction atmospheric conditions, as well as the existing of oxidant elevation should be considered in local air quality control. This study demonstrates the efficacy of ML methods in simulating nonlinear response of O3, supports the understanding of local O3 formation and quick guidance for precise local O3 pollution control and the related strategies.