Search

Microbial ecological functions are an emergent property of community composition. For some ecological functions, this link is strong enough that community composition can be used to estimate the quantity of an ecological function. Here, we apply random forest regression models to compare the predictive performance of community composition and environmental data for bacterial production (BP). Using data from two independent long-term ecological research sites-Palmer LTER in Antarctica and Station SPOT in California-we found that community composition was a strong predictor of BP. The top performing model achieved an R2 of 0.84 and RMSE of 20.2 pmol L-1 hr-1 on independent validation data, outperforming a model based solely on environmental data (R2 = 0.32, RMSE = 51.4 pmol L-1 hr-1). We then operationalized our top performing model, estimating BP for 346 Antarctic samples from 2015 to 2020 for which only community composition data were available. Our predictions resolved spatial trends in BP with significance in the Antarctic (P value = 1 × 10-4) and highlighted important taxa for BP across ocean basins. Our results demonstrate a strong link between microbial community composition and microbial ecosystem function and begin to leverage long-term datasets to construct models of BP based on microbial community composition.

Introduction: Many patients used telehealth services during the COVID-19 pandemic. In this study, we evaluate how different factors have affected telehealth utilization in recent years. Decision makers at the federal and state levels can use the results of this study to inform their healthcare-related policy decisions. Methods: We implemented data analytics techniques to determine the factors that explain the use of telehealth by developing a case study using data from Arkansas. Specifically, we built a random forest regression model which helps us identify the important factors in telehealth utilization. We evaluated how each factor impacts the number of telehealth patients in Arkansas counties. Results: Of the 11 factors evaluated, five are demographic, and six are socioeconomic factors. Socioeconomic factors are relatively easier to influence in the short term. Based on our results, broadband subscription is the most important socioeconomic factor and population density is the most important demographic factor. These two factors were followed by education level, computer use, and disability in terms of their importance as it relates to telehealth use. Discussion: Based on studies in the literature, telehealth has the potential to improve healthcare services by improving doctor utilization, reducing direct and indirect waiting times, and reducing costs. Thus, federal and state decision makers can influence the utilization of telehealth in specific locations by focusing on important factors. For example, investments can be made to increase broadband subscriptions, education levels, and computer use in targeted locations. [ABSTRACT FROM AUTHOR]

In this paper, a digital twin of the network of heating systems for smart cities is developed using the example of the city of Almaty. The study used machine learning algorithms to estimate future thermal energy consumption and develop thermodynamic formulas. This work offers a thorough and in-depth analysis of thermal energy consumption. In addition, the paper identifies the relationship between thermal energy consumption and ambient temperature, and wind uncertainty in certain urban areas using machine learning methods to predict thermal energy consumption. Using both training and regression models, this interdependence is revealed. The obtained forecasts provide useful information for studying the structure of heat consumption in Almaty and reducing heat losses by reducing overheating in the zones of heating networks. In addition, the study analyzes high-resolution spatial data collected from 385 homes and 62 heat transfer circuits located throughout the city during the heating season. The study examines the degree of relationship between the ambient temperature and the amount of heat energy used in the areas of Astana. A minor impact of wind speed is also estimated. These discoveries allow us to use machine learning algorithms to find the location of hot spots and inefficient zones with high losses. [ABSTRACT FROM AUTHOR]

Forecasting future citations based on trends in A.I. publications in libraries is essential for understanding the long-term impact of academic research. This study evaluates how well historical citation data from 2003 to 2023 can predict future citation patterns and compares the effectiveness of various forecasting models. The analysis includes traditional time series models like ARIMA and Exponential Smoothing, as well as machine learning techniques such as Linear Regression, Decision Tree Regression, Random Forest Regression, and Gradient Boosting Regression. ARIMA and Exponential Smoothing are chosen for their robustness in handling historical trends, while machine learning models are applied to capture complex, nonlinear relationships. The study assesses model performance using metrics such as Mean Squared Error (MSE) and R-squared values, identifying the most accurate models for forecasting. The findings reveal that Exponential Smoothing provides reliable forecasts of future citations, while Random Forest Regression demonstrates superior performance among machine learning techniques. Additionally, linear regression models predict citation and publication trends effectively, though they may benefit from more sophisticated approaches for greater accuracy. This research offers valuable insights for researchers, institutions, and policymakers, helping them anticipate the future impact of academic publications and guiding strategic decisions in research and funding. [ABSTRACT FROM AUTHOR]

Accurate mapping of above-ground biomass (AGB) is essential for carbon stock quantification and climate change impact assessment, particularly in mountainous areas. This study applies a random forest (RF) regression model to predict the spatial distribution of AGB in Usho (site A) and Utror (site B) forests located in the northern mountainous region of Pakistan. The predicted maps elucidate AGB variations across these sites, with non-forest areas excluded based on an normalized difference vegetation index (NDVI) threshold value of <0.4. Three different combinations of input datasets were used to predict the biomass, including spectral bands (SBs) only, vegetation indexes (VIs) only, and a combination of both spectral bands and vegetation indexes (SBVIs). Utilizing SBs, the biomass ranged between 150 and 286 mg/ha in site A and 99 and 376 mg/ha in site B. Meanwhile, using VIs indicated a biomass range of 163 Mg/ha–337 Mg/ha and 131–392 Mg/ha for sites A and B, respectively. The combination of spectral bands and vegetation indexes yielded AGB values of 145–290 Mg/ha in site A and 116–389 Mg/ha in site B. The northern and western regions of site A, characterized by higher altitudes and lower forest density, notably showed lower biomass values than other regions. Conversely, similar regions in site B, situated at lower latitudes, demonstrated different biomass ranges. The RF model exhibited robust accuracy, with R2 values of 0.74 and 0.83 for spectral bands and vegetation indexes, respectively. However, with a combination of both, an R2 of 0.79 was achieved. Furthermore, altitudinal gradients significantly influence the biomass distribution across both sites, with specific elevation ranges yielding optimal results. The AGB variation along the slope further corroborated these findings. In both sites, the western aspects showed the highest biomass across all combinations of input datasets. The variable importance analysis highlighted that ARVI8a, NDI45, Band12, Band11, TSAVI8, and ARVI8a are significant predictors in sites A and B. This comprehensive analysis enhances our understanding of AGB distribution in the mountainous forests of Pakistan, offering valuable insights for forest management and ecological studies. [ABSTRACT FROM AUTHOR]

To implement a fully data driven strategy for identifying outliers in clinical datasets used for formula constant optimisation, in order to achieve proper formula predicted refraction after cataract surgery, and to assess the capabilities of this outlier detection method. 2 clinical datasets (DS1/DS2: N = 888/403) of eyes treated with a monofocal aspherical intraocular lens (Hoya XY1/Johnson&Johnson Vision Z9003) containing preoperative biometric data, power of the lens implant and postoperative spherical equivalent (SEQ) were transferred to us for formula constant optimisation. Original datasets were used to generate baseline formula constants. A random forest quantile regression algorithm was set up using bootstrap resampling with replacement. Quantile regression trees were grown and the 25% and 75% quantile, and the interquartile range were extracted from SEQ and formula predicted refraction REF for the SRKT, Haigis and Castrop formulae. Fences were defined from the quantiles and data points outside the fences were marked and removed as outliers before recalculating the formula constants. N B = 1000 bootstrap samples were derived from both datasets, and random forest quantile regression trees were grown to model SEQ versus REF and to estimate the median and 25% and 75% quantiles. The fence boundaries were defined as being from 25% quantile - 1.5·IQR to 75% quantile + 1.5·IQR, with data points outside the fence being marked as outliers. In total, for DS1 and DS2, 25/27/32 and 4/5/4 data points were identified as outliers for the SRKT/Haigis/Castrop formulae respectively. The respective root mean squared formula prediction errors for the three formulae were slightly reduced from: 0.4370 dpt;0.4449 dpt/0.3625 dpt;0.4056 dpt/and 0.3376 dpt;0.3532 dpt to: 0.4271 dpt;0.4348 dpt/0.3528 dpt;0.3952 dpt/0.3277 dpt;0.3432 dpt for DS1;DS2. We were able to prove that with random forest quantile regression trees a fully data driven outlier identification strategy acting in the response space is achievable. In a real life scenario this strategy has to be complemented by an outlier identification method acting in the parameter space for a proper qualification of datasets prior to formula constant optimisation. [ABSTRACT FROM AUTHOR]

This study employs machine learning techniques to forecast horticultural production in Indramayu Regency, Indonesia, utilizing data from the Indramayu Regency Statistics Agency from 2009 to 2017. The variables under observation encompass mango fruit production volume, harvest area, rainfall, and the number of rainy days. Mango fruit production volume is the target variable, while the remaining data serves as features. Regression models comprise Linear Regression, Random Forest Regression, Gradient Boosting Regression, and Decision Tree Regression. The research unveils three key findings. Firstly, it underscores the significance of data preprocessing to eliminate noise or outliers, thereby enhancing the performance of regression models, as evidenced by amplified R2 and reduced RMSE values alongside diminished MAPE. Elevated RMSE values highlight the presence of noise or outliers in unprocessed data. Secondly, it emphasizes the necessity of a representative test data proportion for precise prediction outcomes, as indicated by escalating MAPE and RMSE with increased test data proportions. Lastly, it shows the strong correlation between harvest area and mango production volume, culminating in commendable evaluation metrics. Among the regression models, Random Forest Regression emerges as the most robust, boasting the highest R2 value and lowest RMSE, affirming its efficacy in this study. [ABSTRACT FROM AUTHOR]

The measurement of the macronutrient values of an oil palm plantation is a complex and tedious task, particularly when dealing with large plantation areas. This situation complicates the process of the conventional measurement of nutrients by taking samples of oil palm leaves in the area being observed, causing delays in fertilizer recommendation and a lack of spatial diversity observation. Precision agriculture (PA) principles and approaches, which focus on assessing temporal and spatial variability, can be used to improve conventional measurement methods in terms of both accuracy and speed. This research aims to determine macronutrients, specifically nitrogen (N), phosphorus (P), and potassium (K) contents in oil palm leaves based on PA principles using the integration of remote sensing technology and machine learning to quickly obtain the macronutrient status from oil palm plantation areas. The Sentinel‐1A and Sentinel‐2A imagery data were analyzed and extracted to produce selected features, which are most influencing in the correlation between the imagery data and the leaf macronutrient values obtained from laboratory analysis. The random forest regression (RFR) model is used to produce correlation functions to compute macronutrient values. The use of the two satellites is to cope with cloud and smoke interference. The prototype system developed, named PreciPalm (Precision Agriculture Platform for Oil Palm), has been validated and implemented based on 2000 leaf sampling units representing several oil palm plantation areas in Indonesia, including Java, Sumatra, and Kalimantan. The observed system performance resulted in the measurement accuracy of 95.02%, 93.50%, and 82.52% for the nutrients N, P, and K, respectively. The novelty of PreciPalm is that it provides an ecosystem to transparently measure and observe the macronutrient status of oil palm in a timely, visual, spatial, and location‐specific manner, thereby improving oil palm nutritional management with more certainty and precision. [ABSTRACT FROM AUTHOR]

Introduction: Extreme weather becomes increasingly frequent and severe under climate change, causing unexpected damage to trees. Among them, sleet damage is particularly harmful to evergreen trees in subtropical area. Camphor trees (Cinnamomum camphora), as dominant street trees in central China, are prone to sleet damage, resulting in loss of valuable ecosystem functions. Methods: By measuring tree size characteristics of 118 camphor trees before and after a record-breaking sleet event in Wuhan, a mega-city in central China, we built allometric equations between size and volume of broken branches and used the random forest regression to model sleet damage to camphor trees. Results: We identified that larger trees with intermediate bole height suffered more than smaller trees with tall bole height from the sleet event. We estimated the volume of broken branches of a camphor tree with DBH at 35.0 cm as 106.4 dm3, equivalent to 55.3 kg biomass loss, from the sleet event. Discussion: We suggest that pruning the branches instead of topping the main stems of small camphor trees would reduce the sleet hazard. To mitigate the negative impacts of climate change, regular pruning should be practiced on street camphor trees to protect them from future heavy sleet events. [ABSTRACT FROM AUTHOR]

Urban trees play an important role in mitigating effects of climate change and provide essential ecosystem services. However, the urban environment can stress trees, requiring the use of effective monitoring methods to assess their health and functionality. The objective of this study, which focused on four deciduous tree species in Rennes, France, was to evaluate the ability of hybrid inversion models to estimate leaf chlorophyll content (LCC), leaf area index (LAI), and canopy chlorophyll content (CCC) of urban trees using eight Sentinel-2 (S2) images acquired in 2021. Simulations were performed using the 3D radiative transfer model DART, and the hybrid inversion models were developed using machine-learning regression algorithms (random forest (RF) and gaussian process regression). Model performance was assessed using in situ measurements, and relations between satellite data and in situ measurements were investigated using spatial allocation (SA) methods at the pixel and tree scales. The influence of including environment features (EFs) as model inputs was also assessed. The results indicated that random forest models that included EFs and used the pixel-scale SA method were the most accurate with R2 values of 0.33, 0.29, and 0.46 for LCC, LAI, and CCC, respectively, with notable variability among species. [ABSTRACT FROM AUTHOR]

Estimating the spatiotemporal variations in natural grassland carrying capacity is crucial for maintaining the balance between grasslands and livestock. However, accurately assessing this capacity presents significant challenges due to the high costs of biomass measurement and the impact of human activities. In this study, we propose a novel method to estimate grassland carrying capacity based on potential net primary productivity (NPP), applied to the source area of the Nujiang River and Selinco Lake on the Tibetan Plateau. Initially, we utilize multisource remote sensing data—including soil, topography, and climate information—and employ the random forest regression algorithm to model potential NPP in areas where grazing is banned. The construction of the random forest model involves rigorous feature selection and hyperparameter optimization, enhancing the model's accuracy. Next, we apply this trained model to areas with grazing, ensuring a more accurate estimation of grassland carrying capacity. Finally, we analyze the spatiotemporal variations in grassland carrying capacity. The main results showed that the model achieved a high level of precision, with a root mean square error (RMSE) of 4.89, indicating reliable predictions of grassland carrying capacity. From 2001 to 2020, the average carrying capacity was estimated at 9.44 SU/km2, demonstrating a spatial distribution that decreases from southeast to northwest. A slight overall increase in carrying capacity was observed, with 65.7% of the area exhibiting an increasing trend, suggesting that climate change has a modest positive effect on the recovery of grassland carrying capacity. Most of the grassland carrying capacity is found in areas below 5000 m in altitude, with alpine meadows and alpine meadow steppes below 4750 m being particularly suitable for grazing. Given that the overall grassland carrying capacity remains low, it is crucial to strictly control local grazing intensity to mitigate the adverse impacts of human activities. This study provides a solid scientific foundation for developing targeted grassland management and protection policies. [ABSTRACT FROM AUTHOR]

The main objective of this paper is to assess the vulnerability of reinforced concrete (RC) educational buildings in Dhaka city to seismic activity by utilizing machine learning (ML) algorithms. There are three main stages in traditional seismic vulnerability assessment: rapid visual assessment (RVA), preliminary engineering assessment (PEA), and detailed engineering assessment (DEA). The conventional three‐step evaluation process for determining the seismic vulnerability of existing buildings is time‐consuming and expensive, especially when dealing with a large building stock or a city. This study focuses on using an ML‐based approach to evaluate seismic vulnerability, specifically in terms of the story shear ratio (SSR), which serves as the risk index. The main concept is the utilization of RVA data to obtain analytical results (SSR). The dataset utilized in this study comprises RVA data for 268 buildings and corresponding PEA data for the same 268 buildings. The RVA data include the construction year, condition, typical floor area, number of stories, total floor area, additions, alterations, redundancy, pounding, and irregularities. The PEA data comprise SSR, which was generated from linear dynamic analysis. These data were collected from the Urban Resilience Project of Rajdhani Unnayan Kartripakkha (RAJUK), which is the development authority of Dhaka. Random forest regression (RFR), support vector regression (SVR), and artificial neural networks (ANNs) are employed to determine the SSR of existing educational RC buildings. A comparative analysis for each model is also made. From the analysis results, it shows that RFR, ANN, and SVR achieved coefficient of determination (R2) of 20%, 25%, and 35%, respectively. Based on the findings from the three separate model analyses, it can be concluded that SVR produced the highest performance among the considered models. [ABSTRACT FROM AUTHOR]

Low carbon management and policies should refer to local long-term inter-annual carbon uptake. However, most previous research has only focused on the quantity and spatial distribution of gross primary product (GPP) for the past 50 years because most satellite launches, the main GPP data source, were no earlier than 1980. We identified a close relationship between the tree-ring index (TRI) and vegetation carbon dioxide uptake (as measured by GPP) and then developed a nested TRI-GPP model to reconstruct spatially explicit GPP values since 1895 from seven tree-ring chronologies. The model performance in both phases was acceptable: We chose general regression neural network regression and random forest regression in Phase 1 (1895–1937) and Phase 2 (1938–1985). With the simulated and real GPP maps, we observed that the GPP for grassland and overall GPP were increasing. The GPP landscape patterns were stable, but in recent years, the GPP's increasing rate surpassed any other period in the past 130 years. The main local climate driver was the Palmer Drought Severity Index (PDSI), and GPP had a significant positive correlation with PDSI in the growing season (June, July, and August). With the GPP maps derived from the nested TRI-GPP model, we can create fine-scale GPP maps to understand vegetation change and carbon uptake over the past century. [ABSTRACT FROM AUTHOR]

Precision glass molding (PGM) is an effective approach to manufacturing infrared chalcogenide glass (ChG) aspherical lens with complex shapes. However, infrared ChG aspherical lens often experiences form error in the designed profile and the final profile obtained by PGM. To reduce the form error of infrared ChG aspherical lens in the PGM process, a form error compensation model based on the random forest (RF) algorithm is proposed. The infrared ChG aspherical lens profile was first machined on an electroless nickel-phosphorus (Ni–P) plating to serve as the mold for PGM. After molding, the profile data of the lens was extracted, and a compensation model based on RF was constructed to optimize the model parameters using the evaluation parameters such as root mean square error (RMSE), coefficient of determination (R2), and out-of-bag (OOB). Finally, the generated compensated profile based on the compensation model was used for the compensation machining of the mold. Through this compensation approach, we have demonstrated a substantial 63.5 % reduction in the form error of the fabricated infrared ChG aspherical lens, decreasing the Peak-to-Valley (PV) value from 1.04 μm to 0.38 μm. • We propose a novel aspherical mold compensation method based on the random forest algorithm, which is applied to precision molding of Infrared Chalcogenide Glass. • Aspherical mold machining is performed on the electroless nickel-phosphorus (Ni–P) surface for precision molding, and the profile data of lens are extracted for construct the mold compensation machining model. • Through this compensation approach, we have demonstrated a substantial 63.5 % reduction in the form error of the fabricated infrared Chalcogenide Glass aspherical lens, decreasing the Peak-to-Valley (PV) value from 1.04 μm to 0.38 μm. [ABSTRACT FROM AUTHOR]

The objective of this study was to analyse the potential impact of future climate change on grassland cover in Burkina Faso. MODIS NDVI 250 m time series were used to monitor changes in grassland over 2000–2022. The random forest regression (RFR) model was fit by regressing reference data of grassland cover against current climatic and other environmental predictors to predict the current grassland cover map (2022). Projected climate model data of CMIP6 used under SSP 126 and SSP 585 scenarios were integrated into the fit RFR model to predict future change. The results revealed that grassland areas were largely dominated by non-significant productivity change (55%) during 2000–2022. In this period, grassland area knew more increased productivity (35%) than decrease (10%). Burkina Faso is predicted to face more decreased areas of grassland cover than increase by 2061–2080 under SSP 126 and SSP 585 scenarios. The findings of this study can help to set up appropriate adaptation measures to combat climate change in Burkina Faso. [ABSTRACT FROM AUTHOR]

Pakistan is among the South Asian countries mostly vulnerable to the negative health impacts of air pollution. In this context, the study aimed to analyze the spatiotemporal patterns of chronic obstructive pulmonary disease (COPD) incidence and its relationship with air pollutants including aerosol absorbing index (AAI), carbon monoxide, sulfur dioxide (SO2), and nitrogen dioxide. Spatial scan statistics were employed to identify temporal, spatial, and spatiotemporal clusters of COPD. Generalized linear regression (GLR) and random forest (RF) models were utilized to evaluate the linear and non-linear relationships between COPD and air pollutants for the years 2019 and 2020. The findings revealed three spatial clusters of COPD in the eastern and central regions, with a high-risk spatiotemporal cluster in the east. The GLR identified a weak linear relationship between the COPD and air pollutants with R2 = 0.1 and weak autocorrelation with Moran's index = −0.09. The spatial outcome of RF model provided more accurate COPD predictions with improved R2 of 0.8 and 0.9 in the respective years and a very low Moran's I = −0.02 showing a random residual distribution. The RF findings also suggested AAI and SO2 to be the most contributing predictors for the year 2019 and 2020. Hence, the strong association of COPD clusters with some air pollutants highlight the urgency of comprehensive measures to combat air pollution in the region to avoid future health risks. [ABSTRACT FROM AUTHOR]

Although low-cost air quality sensors facilitate the implementation of denser air quality monitoring networks, enabling a more realistic assessment of individual exposure to airborne pollutants, their sensitivity to multifaceted field conditions is often overlooked in laboratory testing. This gap was addressed by introducing an in-field calibration and validation of three PAQMON 1.0 mobile sensing low-cost platforms developed at the Mining and Metallurgy Institute in Bor, Republic of Serbia. A configuration tailored for monitoring PM2.5 and PM10 mass concentrations along with meteorological parameters was employed for outdoor measurement campaigns in Bor, spanning heating (HS) and non-heating (NHS) seasons. A statistically significant positive linear correlation between raw PM2.5 and PM10 measurements during both campaigns (R > 0.90, p ≤ 0.001) was observed. Measurements obtained from the uncalibrated NOVA SDS011 sensors integrated into the PAQMON 1.0 platforms exhibited a substantial and statistically significant correlation with the GRIMM EDM180 monitor (R > 0.60, p ≤ 0.001). The calibration models based on linear and Random Forest (RF) regression were compared. RF models provided more accurate descriptions of air quality, with average adjR2 values for air quality variables in the range of 0.70 to 0.80 and average NRMSE values between 0.35 and 0.77. RF-calibrated PAQMON 1.0 platforms displayed divergent levels of accuracy across different pollutant concentration ranges, achieving a data quality objective of 50% during both measurement campaigns. For PM2.5, uncertainty ( U r ) was below 50% for concentrations between 9.06 and 34.99 μg/m3 in HS and 5.75 and 17.58 μg/m3 in NHS, while for PM10, it stayed below 50% from 19.11 to 51.13 μg/m3 in HS and 11.72 to 38.86 μg/m3 in NHS. [ABSTRACT FROM AUTHOR]

Understanding aboveground biomass (AGB) and its spatial distribution is key to evaluating the productivity and carbon sink effect of forest ecosystems. In this study, a 123-year-old Chinese fir forest in the Dabieshan Mountains of western Anhui Province was used as the research subject. Using AGB data calculated from field measurements of individual Chinese fir trees (diameter at breast height [DBH] and height) and spectral vegetation indices derived from unmanned aerial vehicle (UAV) remote sensing images, a random forest regression model was developed to predict individual tree AGB. This model was then used to estimate the AGB of individual Chinese fir trees. Combined with digital elevation model (DEM) data, the effects of topographic factors on the spatial distribution of AGB were analyzed. We found that remote sensing spectral vegetation indices obtained by UAVs can be used to predict the AGB of individual Chinese fir trees, with the normalized difference vegetation index (NDVI) and the optimized soil-adjusted vegetation index (OSAVI) being two important predictors. The estimated AGB of individual Chinese fir trees was 339.34 Mg·ha−1 with a coefficient of variation of 23.21%. At the local scale, under the influence of elevation, slope, and aspect, the AGB of individual Chinese fir trees showed a distribution pattern of decreasing from the middle to the northwest and southeast along the northeast-southwest trend. The effect of elevation on AGB was influenced by slope and aspect; AGB on steep slopes was higher than on gentle slopes, and the impact of slope on AGB was influenced by aspect. Additionally, AGB on north-facing slopes was higher than on south-facing slopes. Our results suggest that local environmental factors such as elevation, slope, and aspect should be considered in future Chinese fir plantation management and carbon sink assessments in the Dabieshan Mountains of western Anhui, China. [ABSTRACT FROM AUTHOR]

Purpose: To implement a fully data driven strategy for identifying outliers in clinical datasets used for formula constant optimisation, in order to achieve proper formula predicted refraction after cataract surgery, and to assess the capabilities of this outlier detection method. Methods: 2 clinical datasets (DS1/DS2: N = 888/403) of eyes treated with a monofocal aspherical intraocular lens (Hoya XY1/Johnson&Johnson Vision Z9003) containing preoperative biometric data, power of the lens implant and postoperative spherical equivalent (SEQ) were transferred to us for formula constant optimisation. Original datasets were used to generate baseline formula constants. A random forest quantile regression algorithm was set up using bootstrap resampling with replacement. Quantile regression trees were grown and the 25% and 75% quantile, and the interquartile range were extracted from SEQ and formula predicted refraction REF for the SRKT, Haigis and Castrop formulae. Fences were defined from the quantiles and data points outside the fences were marked and removed as outliers before recalculating the formula constants. Results: NB = 1000 bootstrap samples were derived from both datasets, and random forest quantile regression trees were grown to model SEQ versus REF and to estimate the median and 25% and 75% quantiles. The fence boundaries were defined as being from 25% quantile - 1.5·IQR to 75% quantile + 1.5·IQR, with data points outside the fence being marked as outliers. In total, for DS1 and DS2, 25/27/32 and 4/5/4 data points were identified as outliers for the SRKT/Haigis/Castrop formulae respectively. The respective root mean squared formula prediction errors for the three formulae were slightly reduced from: 0.4370 dpt;0.4449 dpt/0.3625 dpt;0.4056 dpt/and 0.3376 dpt;0.3532 dpt to: 0.4271 dpt;0.4348 dpt/0.3528 dpt;0.3952 dpt/0.3277 dpt;0.3432 dpt for DS1;DS2. Conclusion: We were able to prove that with random forest quantile regression trees a fully data driven outlier identification strategy acting in the response space is achievable. In a real life scenario this strategy has to be complemented by an outlier identification method acting in the parameter space for a proper qualification of datasets prior to formula constant optimisation.

Abstract Modeling implied volatility (IV) is important for option pricing, hedging, and risk management. Previous studies of deterministic implied volatility functions (DIVFs) propose two parameters, moneyness and time to maturity, to estimate implied volatility. Recent DIVF models have included factors such as a moving average ratio and relative bid-ask spread but fail to enhance modeling accuracy. The current study offers a generalized DIVF model by including a momentum indicator for the underlying asset using a relative strength index (RSI) covering multiple time resolutions as a factor, as momentum is often used by investors and speculators in their trading decisions, and in contrast to volatility, RSI can distinguish between bull and bear markets. To the best of our knowledge, prior studies have not included RSI as a predictive factor in modeling IV. Instead of using a simple linear regression as in previous studies, we use a machine learning regression algorithm, namely random forest, to model a nonlinear IV. Previous studies apply DVIF modeling to options on traditional financial assets, such as stock and foreign exchange markets. Here, we study options on the largest cryptocurrency, Bitcoin, which poses greater modeling challenges due to its extreme volatility and the fact that it is not as well studied as traditional financial assets. Recent Bitcoin option chain data were collected from a leading cryptocurrency option exchange over a four-month period for model development and validation. Our dataset includes short-maturity options with expiry in less than six days, as well as a full range of moneyness, both of which are often excluded in existing studies as prices for options with these characteristics are often highly volatile and pose challenges to model building. Our in-sample and out-sample results indicate that including our proposed momentum indicator significantly enhances the model’s accuracy in pricing options. The nonlinear machine learning random forest algorithm also performed better than a simple linear regression. Compared to prevailing option pricing models that employ stochastic variables, our DIVF model does not include stochastic factors but exhibits reasonably good performance. It is also easy to compute due to the availability of real-time RSIs. Our findings indicate our enhanced DIVF model offers significant improvements and may be an excellent alternative to existing option pricing models that are primarily stochastic in nature.

Traffic prediction is essential for transportation planning, resource allocation, congestion management and enhancing travel experiences. This study optimizes data preprocessing techniques to improve machine learning-based traffic prediction models. Data preprocessing is critical in preparing the data for machine learning models. This study proposes an approach that optimizes data preprocessing techniques, focusing on flow-based analysis and optimization, to enhance traffic prediction models. The proposed approach explores fixed and variable orders of data preprocessing using a genetic algorithm across five diverse datasets. Evaluation metrics such as root mean squared error (RMSE), mean absolute error (MAE) and R-squared error assess model performance. The results indicate that the genetic algorithm’s variable order achieves the best performance for the ArcGIS Hub and Frementon Bridge Cycle datasets, fixed order one preprocessing for the Traffic Prediction dataset and variable order using the genetic algorithm for the PeMS08 dataset. Fixed order 2 preprocessing yields the best performance for the XI AN Traffic dataset. These findings highlight the importance of selecting the appropriate data preprocessing flow order for each dataset, improving traffic prediction accuracy and reliability. The proposed approach advances traffic prediction methodologies, enabling more precise and reliable traffic forecasts for transportation planning and management applications. [ABSTRACT FROM AUTHOR]

Copyright of Acta Materiae Compositae Sinica is the property of Acta Materiea Compositae Sinica Editorial Department and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Surgeons face a significant challenge due to the heat generated during drilling, as excessive temperatures at the bone–tool interface can lead to irreversible damage to the regenerative soft tissue and result in thermal osteonecrosis. While previous studies have explored the use of machine learning to predict the temperature rise during bone drilling, this in vitro study introduces a comprehensive approach by combining the Response Surface Methodology (RSM) with advanced machine learning techniques. The main objective lies in the comprehensive evaluation and comparison of support vector machine (SVM) and random forest (RF) models specifically for the optimization of the bone drilling parameters to prevent thermal bone necrosis. A total of 27 experiments were conducted using a multi-level factorial method, with analysis performed via the Minitab software version 19.1. Performance metrics such as the mean squared error (MSE), mean absolute percentage error (MAPE), and coefficient of determination (R2) were used to assess model accuracy. The RF model emerged as the most effective, with R2 values of 94.2% for testing and 97.3% for training data, significantly outperforming other models in predicting temperature fluctuations. This study demonstrates the superior predictive capabilities of the RF model and offers a robust framework for the optimization of surgical procedures to mitigate the risk of thermal damage. [ABSTRACT FROM AUTHOR]

This paper provides a thorough comparative review of deception detection techniques employed for a sample of 400 convicted offenders. It focuses on the utilization of polygraph sensor data as input variables for predicting deception, which are assessed against manual scoring by experts. Three advanced machine learning models, namely Random Forest Regression (RFR), Support Vector Machine (SVM) Regression, and Neural Network Regression (NNR), were employed with the purpose of analysing their predictive efficacy in identifying deception based on physiological responses captured by polygraph sensors. The obtained results indicate that all three algorithms exhibited varying degrees of effectiveness in predicting deceptive behavior. The Random Forest Regression algorithm achieved a Mean Squared Error (MSE) of 0.893 and a coefficient of determination (R²) of 0.091, which highlights its ability to discern key physiological indicators related to deceptive behavior. The Support Vector Machine Regression algorithm showed a competitive performance with a MSE of 0.98 and a R² value of 0.159, which underscores its capability to model non-linear relationships in the context of high-dimensional data. However, the Neural Network Regression algorithm proved to be the best model, with a MSE of 0.894 and a significantly higher R² value of 0.113. This model’s capacity to capture the complex relationships in the context of physiological data allowed it to surpass both RFR and SVM, which indicates its potential for a precise and reliable deception detection. This study provides valuable insights into the advancement of forensic applications with regard to deception detection technologies. Its findings suggest that the Neural Network Regression algorithm, due to its ability to learn complex patterns and relationships related to physiological data, stands out as an optimal choice for accurately identifying deceptive behavior. [ABSTRACT FROM AUTHOR]

Mountains have a critical role in freshwater supply for downstream populations. As the climate changes, groundwater stored in mountains may help buffer the impacts to declining water resources caused by decreased snowpack and glacier recession. However, given the scarcity of groundwater observation wells in mountain regions, it remains unclear how mountain groundwater is being impacted by climate change across ecoregions. This study quantifies temporal trends in mountain groundwater levels and explores how various climatic, physiographic and anthropogenic factors affect these trends. We compiled data from 171 public groundwater observation wells within mountain regions across Canada and the United States, for which at least 20 years of monthly data is available. The Mann‐Kendall test for monotonic trend revealed that 54% of these wells have statistically significant temporal trends (p < 0.05) over the period of record, of which 69% were negative and therefore indicating overall declining groundwater storage. Wells in the western mountain ranges showed stronger trends (both positive and negative) than the eastern mountain ranges, and higher elevation wells showed fewer negative trends than the low elevation (<400 m asl) wells (p < 0.05). Correlation, Kruskal‐Wallis tests, stepwise multiple linear regression and random forest regression were used to identify factors controlling groundwater trends. Statistical analysis revealed that lower‐elevation mountain regions with higher average annual temperatures and lower average annual precipitation have the greatest declines in groundwater storage under climate change. Trends in temperature and precipitation, and ecoregion were also important predictors on groundwater level trends, highlighting geographic differences in how mountain wells are responding to climate change. Furthermore, sedimentary bedrock aquifers showed markedly more negative trends than crystalline bedrock aquifers. The findings demonstrate that the impact of climate change on mountain water resources extends to the subsurface, with important implications for global water resources. [ABSTRACT FROM AUTHOR]

Modeling implied volatility (IV) is important for option pricing, hedging, and risk management. Previous studies of deterministic implied volatility functions (DIVFs) propose two parameters, moneyness and time to maturity, to estimate implied volatility. Recent DIVF models have included factors such as a moving average ratio and relative bid-ask spread but fail to enhance modeling accuracy. The current study offers a generalized DIVF model by including a momentum indicator for the underlying asset using a relative strength index (RSI) covering multiple time resolutions as a factor, as momentum is often used by investors and speculators in their trading decisions, and in contrast to volatility, RSI can distinguish between bull and bear markets. To the best of our knowledge, prior studies have not included RSI as a predictive factor in modeling IV. Instead of using a simple linear regression as in previous studies, we use a machine learning regression algorithm, namely random forest, to model a nonlinear IV. Previous studies apply DVIF modeling to options on traditional financial assets, such as stock and foreign exchange markets. Here, we study options on the largest cryptocurrency, Bitcoin, which poses greater modeling challenges due to its extreme volatility and the fact that it is not as well studied as traditional financial assets. Recent Bitcoin option chain data were collected from a leading cryptocurrency option exchange over a four-month period for model development and validation. Our dataset includes short-maturity options with expiry in less than six days, as well as a full range of moneyness, both of which are often excluded in existing studies as prices for options with these characteristics are often highly volatile and pose challenges to model building. Our in-sample and out-sample results indicate that including our proposed momentum indicator significantly enhances the model's accuracy in pricing options. The nonlinear machine learning random forest algorithm also performed better than a simple linear regression. Compared to prevailing option pricing models that employ stochastic variables, our DIVF model does not include stochastic factors but exhibits reasonably good performance. It is also easy to compute due to the availability of real-time RSIs. Our findings indicate our enhanced DIVF model offers significant improvements and may be an excellent alternative to existing option pricing models that are primarily stochastic in nature. [ABSTRACT FROM AUTHOR]

Laser powder bed fusion (L-PBF) fuses metallic powder using a high-energy laser beam, forming parts layer by layer. This technique offers flexibility and design freedom in metal additive manufacturing (MAM). However, achieving the desired surface quality remains challenging and impacts functionality and reliability. L-PBF process parameters significantly influence surface roughness. Identifying the most critical factors among numerous parameters is essential for improving quality. This study examines the effects of key process parameters on the surface roughness of AlSi10Mg, a widely used aluminum alloy in high-tech industries, fabricated by L-PBF. Part orientation, laser power, scanning speed, and layer thickness were identified as crucial parameters via cause-and-effect analysis. To systematically examine their effects, the Taguchi method was employed within the framework of the design of experiment (DoE). Experimental results and statistical analysis revealed that laser power, scanning speed, and layer thickness significantly influence surface roughness parameters: arithmetic mean (Ra) and root mean square (Rq). Main effect plots and energy density analyses confirmed their impact on surface quality. Microscopic investigations identified surface flaws such as spattering, balling, and porosity contributing to poor quality. Given the complex interplay between parameters and surface quality, accurately predicting their effects is challenging. To address this, machine learning models, specifically random forest regression (RFR) and support vector regression (SVR), were used to predict the effects on surface roughness. The RFR model's R2 values for predicting Ra and Rq are 97% and 85%, while the SVR model's predictions are 85% and 66%, respectively. Evaluation metrics demonstrated that the RFR model outperformed SVR in predicting surface roughness. [ABSTRACT FROM AUTHOR]

Machine learning-based prediction of the output characteristics of a negative capacitance (NC) double gate (DG) Tunnel FET (TFET) (NC DG TFET) has been demonstrated in this paper. On current (ION), OFF current (IOFF) and sub-threshold slope (SS) have been predicted for different values of the body thickness (TSi), oxide thickness (TOX) and channel length (LCh) and the work function (WF). Random Forest Regression (RFR) model is used in the ML framework to explore the device characteristics. A total of 5444 simulated data has been generated by using the SILVACO ATLAS tool. After pre-processing the data, RFR-ML has been applied to obtain maximum accuracy. Root means square error (RMSE) (~0.34), R2 score (~0.965) and accuracy (~98.51%) parameter have been used as the figures of merit (FOM) to judge the efficiency of the proposed model with respect to the state-of-the-art literature. Besides, with higher accuracy, the model is capable of predicting the data within a few seconds compared to the few hundred hours using TCAD tools. This fast and accurate design strategy will surely be efficacious for the present VLSI industries. [ABSTRACT FROM AUTHOR]

This article aims to evaluate the performance of solar radiation forecasts produced by CMA-WSP v2.0 (version 2 of the China Meteorological Administration Wind and Solar Energy Prediction System) and to explore the application of machine learning algorithms from the scikit-learn Python library to improve the solar radiation prediction made by the CMA-WSP v2.0. It is found that the performance of the solar radiation forecasting from the CMA-WSP v2.0 is closely related to the weather conditions, with notable diurnal fluctuations. The mean absolute percentage error (MAPE) produced by the CMA-WSP v2.0 is approximately 74% between 11:00 and 13:00. However, the MAPE ranges from 193% to 242% at 07:00–08:00 and 17:00–18:00, which is greater than that observed at other daytime periods. The MAPE is relatively low (high) for both sunny and cloudy (overcast and rainy) conditions, with a high probability of an absolute percentage error below 25% (above 100%). The forecasts tend to underestimate (overestimate) the observed solar radiation in sunny and cloudy (overcast and rainy) conditions. By applying machine learning models (such as linear regression, decision trees, K-nearest neighbors, random forests regression, adaptive boosting, and gradient boosting regression) to revise the solar radiation forecasts, the MAPE produced by the CMA-WSP v2.0 is significantly reduced. The reduction in the MAPE is closely connected to the weather conditions. The models of K-nearest neighbors, random forests regression, and decision trees can reduce the MAPE in all weather conditions. The K-nearest neighbor model exhibits the most optimal performance among these models, particularly in rainy conditions. The random forest regression model demonstrates the second-best performance compared to that of the K-nearest neighbor model. The gradient boosting regression model has been observed to reduce the MAPE of the CMA-WSP v2.0 in all weather conditions except rainy. In contrast, the adaptive boosting (linear regression) model exhibited a diminished capacity to improve the CMA-WSP v2.0 solar radiation prediction, with a slight reduction in MAPE observed only in sunny (sunny and cloudy) conditions. In addition, the input feature selection has a considerable influence on the performance of the machine learning model. The incorporation of the time series data associated with the diurnal variation of solar radiation as an input feature can further improve the model's performance. [ABSTRACT FROM AUTHOR]

Medical data are subject to privacy regulations, which severely limit AI specialists who wish to construct decision support systems for medicine. Large amounts of this data are tabular, indicating that they are organized into a table format, where patient records are represented in rows and measured variables in columns. Furthermore, the variables come in different types—some are numerical, while others are categorical. In this work, we introduce a novel method for constructing generators of synthetic tabular data with mixed types. The key point of our approach is the explicit utilization of a latent space to represent the original data. A case study using real medical data is presented. [ABSTRACT FROM AUTHOR]

Customer reviews are the second-most reliable source of information, followed by family and friend referrals. However, there are many existing customer reviews. Some online shopping platforms address this issue by ranking customer reviews according to their usefulness. However, we propose an alternative method to rank customer reviews, given that this system is easily manipulable. This study aims to create a ranking model for reviews based on their usefulness by combining product and seller service aspects from customer reviews. This methodology consists of six primary steps: data collection and preprocessing, aspect extraction and sentiment analysis, followed by constructing a regression model using random forest regression, and the review ranking process. The results demonstrate that the ranking model with service considerations outperformed the model without service considerations. This demonstrates the model's superiority in the three tests, which include a comparison of the regression results, the aggregate helpfulness ratio, and the matching score. [ABSTRACT FROM AUTHOR]

Conventional concrete causes significant environmental problems, including resource depletion, high CO2 emissions, and high energy consumption. Steel slag aggregate (SSA), a by-product of the steelmaking industry, offers a sustainable alternative due to its environmental benefits and improved mechanical properties. This study examined the predictive power of four modeling techniques—Gene Expression Programming (GEP), an Artificial Neural Network (ANN), Random Forest Regression (RFR), and Gradient Boosting (GB)—to predict the compressive strength (CS) of SSA concrete. Using 367 datasets from the literature, six input variables (cement, water, granulated furnace slag, superplasticizer, coarse aggregate, fine aggregate, and age) were utilized to predict compressive strength. The models' performance was evaluated using statistical measures such as the mean absolute error (MAE), root mean squared error (RMSE), mean values, and coefficient of determination (R2). Results indicated that the GB model consistently outperformed RFR, GEP, and the ANN, achieving the highest R2 values of 0.99 and 0.96 for the training and testing dataset, respectively, followed by RFR with R2 values of 0.97 (training) and 0.93 (testing), GEP with R2 values of 0.85 (training) and 0.87 (testing), and ANN with R2 values of 0.61 (training) and 0.82 (testing). Additionally, the GB model had the lowest MAE values of 0.79 MPa (training) and 2.61 MPa (testing) and RMSE values of 1.90 MPa (training) and 3.95 MPa (testing). This research aims to advance predictive modeling in sustainable construction through analysis and well-defined conclusions. [ABSTRACT FROM AUTHOR]

Imphal city and its surrounding regions have been experiencing a wide range of changes in land use and land cover (LULC) over recent years. However, there is a shortage of comprehensive studies that specifically examine the long-term dynamics in general and driving factors behind these changes in particular. This study provides a detailed analysis of LULC changes in Imphal city and its surrounding regions from 1991 to 2021, using advanced geospatial techniques to investigate underlying dynamics and driving factors behind urbanization. Multi-spectral datasets from Landsat 7 and 8 were utilized for four distinct time periods (1991, 2001, 2011, and 2021). The supervised classification method, employing the maximum likelihood classifier method, was used to generate the classified LULC maps in the ERDAS 2014. The classified images produced by this technique were evaluated for accuracy through matrix union using the statistical kappa coefficient and overall accuracy measures. Change detection for the periods 1991–2001, 2001–2011, and 2011–2021 was conducted using matrix union (intersection) to identify apparent changes in various LULC classes. The findings reveal a significant increase in built-up areas, expanding from 2440.35 hectares in 1991 to 8205.98 hectares in 2021, at an annual rate of 192.19 hectares/year. In contrast, agricultural lands decreased from 7484.13 hectares in 1991 to 3585.42 hectares in 2021, declining annually by 129.96 hectares/year. The study utilizes classified supervised images from various years, alongside seven independent spatial variables, to analyze urban growth and expansion using the method of random forests regression technique in Python, validated through receiver operating characteristic curve. Analysis of these seven variables, including distance to roads, proximity to the city centre, and infrastructure development, identifies them as primary drivers of urban expansion. The study concludes that rapid urbanization is significantly reshaping LULC dynamics, underscoring the urgent need for sustainable urban planning and management strategies. [ABSTRACT FROM AUTHOR]

Accurate estimation of porosity is a critical factor in reservoir characterization. This study aims to enhance porosity prediction through the implementation and comparison of various stacking ensemble learning strategies. A dataset comprising 273 points, which consists of well logs and core measurements, was collected from two wells for model development. Four base learners, including Support Vector Regression (SVR), Multi-Layer Perceptron (MLP), Random Forest Regression (RFR), and XGBoost, were trained on this dataset. These models were then integrated using multiple stacking ensemble techniques, such as weighted averaging, Bayesian model averaging, and RFR as a metalearner. Meta-learners were trained on predictions from the base learners, generated through crossvalidation on leave-out data. Performance evaluations of both base and meta learners were conducted on a separate testing dataset using statistical and graphical error analysis. Results indicate that all learners demonstrated robust performance, with weighted averaging outperforming other strategies on testing data. The stacking ensemble approach, particularly through weighted averaging, effectively improved base learner performance on testing data by leveraging individual model strengths and mitigating weaknesses. The findings of this study are valuable for geoscientists and reservoir engineers in achieving accurate reservoir characterization and facilitating exploration activities. [ABSTRACT FROM AUTHOR]

A way to quickly and accurately extract building information from visible light images acquired by drones is a hot issue in remote sensing image processing research. In response to the above problems, this paper combined feature extraction, random forest regression, and improved morphological alternating filtering algorithms on the basis of in-depth research on the morphological building index and enhanced morphological building index. A targeted improvement was made to the building extraction method. The main improvements in this paper are as follows. (1) A method for building enhancement of UAV visible light images based on random forest regression is proposed. By fusing the original visible light band, visible light remote sensing index features, texture features, etc., the feature space is constructed, and no artificial feature selection is required while enhancing the building information in the case of the situation. (2) Aiming at the problem of a lot of noise in the homogeneous area inside the enhanced image of the building, the image is morphologically filtered through EASFs. Finally, in order to verify the effectiveness of the method in the paper, the representative city center building area and the city surrounding building area are selected for extraction experiments. The main building types of the two areas are quite different. In the above two areas, the F1 value of the proposed method is increased by 10.40% and 10.08% compared with the MBI algorithm, and the F1 of the first proposed an improved enhanced morphological building index value is increased by 5.23% and 7.87% compared with the EMBI algorithm. The results show that this method can better extract the building information from the UAV's visible light image. [ABSTRACT FROM AUTHOR]

In this study, due to their flexibility in forecasting, the capabilities of three regression-based machine learning models were explored, specifically random forest regression (RFr), generalized regression neural network (GRNN), and support vector regression (SVR). The above models were assessed for their suitability in modeling daily reference evapotranspiration (ETo), based only on temperature data (Tmin, Tmax, Tmean), by comparing their daily ETo results with those estimated by the conventional FAO 56 PM model, which requires a broad range of data that may not be available or may not be of reasonable quality. The RFr, GRNN, and SVR models were subjected to performance evaluation by using statistical criteria and scatter plots. Following the implementation of the ETo models' comparisons, it was observed that all regression-based machine learning models possess the capability to accurately estimate daily ETo based only on temperature data requirements. In particular, the RFr model outperformed the others, achieving the highest R value of 0.9924, while the SVR and GRNN models had R values of 0.9598 and 0.9576, respectively. Additionally, the RFr model recorded the lowest values in all error metrics. Once these regression-based machine learning models have been successfully developed, they will have the potential to serve as effective alternatives for estimating daily ETo, under current and climate change conditions, when temperature data are available. This information is crucial for effective water resources management and especially for predicting agricultural production in the context of climate change. [ABSTRACT FROM AUTHOR]

Copyright of Coal Science & Technology (0253-2336) is the property of Coal Science & Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The traditional microstructure detecting methods such as metallography and electron backscatter diffraction are destructive to the sample and time-consuming and they cannot meet the needs of rapid online inspection. In this paper, a random forest regression microstructure characterization method based on a laser ultrasound technique is investigated for evaluating the microstructure of a titanium alloy (Ti-6Al-4V). Based on the high correlation between the longitudinal wave velocity of ultrasonic waves, the average grain size of the primary α phase, and the volume fraction of the transformed β matrix of the titanium alloy, and with the longitudinal wave velocity as the input feature and the average grain size of the primary α phase and the volume fraction of the transformed β matrix as the output features, prediction models for the average grain size of the primary α phase and the volume fraction of the transformed β matrix were developed based on a random forest regression. The results show that the mean values of the mean relative errors of the predicted mean grain size of the native α phase and the volume fraction of the transformed β matrix for the six samples in the two prediction models were 11.55% and 10.19%, respectively, and the RMSE and MAE obtained from both prediction models were relatively small, which indicates that the two established random forest regression models have a high prediction accuracy. [ABSTRACT FROM AUTHOR]

Vietnam’s rapid urbanization and economic growth have led to an increase in high-rise buildings, making building subsidence a significant concern. Monitoring subsidence is crucial for ensuring building safety and reducing potential risks. The leveling method is commonly used in Vietnam to monitor subsidence, providing valuable data for predicting future subsidence behavior. However, traditional prediction methods based on mathematical models have limitations in capturing complex subsidence patterns. Machine learning techniques have shown promise in enhancing subsidence prediction accuracy. In this study, we analyze machine learning methods for predicting building subsidence using leveling results in Vietnam. We utilize a dataset from a subsidence monitoring network in Hoa Binh General Hospital and compare the performance of linear regression, decision tree regression, and random forest regression models. Our results show that the decision tree and random forest models produce consistent predicted subsidence values, aligning with the observed stability of the building. In contrast, the linear regression model fails to capture the diminishing nature of subsidence over time. We discuss the implications of these findings and highlight the advantages of machine learning in accurately forecasting subsidence. The study demonstrates the potential of machine learning in revolutionizing subsidence prediction and enhancing the monitoring and management of building stability and structural integrity in Vietnam.

This article uses a random forest regression (RFR) model to analyze wind speed forecasting. Wind energy is one of the more critical potentials in renewable energy sources for producing a clean and safe environment. Accurate and stable wind forecasting is essential to improving the efficiency of wind turbines, guaranteeing the power balance, economic dispatch of power systems and ensuring equipment safety. Previous researchers have attempted to address these issues of less wind prediction performance and lack of interpretable analysis. This study aims to develop machine learning (ML) models, such as neural networks (NNs), linear regression (LR), support vector regression (SVR), decision tree regression (DTR), K-nearest neighbors (K-NN), extreme gradient boosting regression and RFR. Six evaluation criteria are applied to estimate the efficiency of the ML model: mean squared error, root mean squared error, mean absolute error (MAE), mean absolute percentage error, normalized average squares of the error and coefficient of determination. The experimental results show the RFR model achieves better prediction accuracy than other models. The forecasting accuracy of the RFR model from wind speed was NMSE = 0.003, MAE = 0.049, MSE = 0.033, RMSE = 0.182, MAPE = 1.180 and R2 = 0.996. Precise wind speed predictions are essential for various industries, such as aviation, shipping and wind energy generation.

Accurate Gross Domestic Product (GDP) prediction is essential for economic planning and policy formulation. This paper evaluates the performance of three machine learning models—Random Forest Regression (RFR), XGBoost, and Prophet—in predicting Somalia's GDP. Historical economic data, including GDP per capita, population, inflation rate, and current account balances, were used in training and testing. Among the models, RFR achieved the best accuracy with the lowest MAE (0.6621%), MSE (1.3220%), RMSE (1.1497%), and R-squared of 0.89. The Diebold-Mariano p-value for RFR (0.042) confirmed its higher predictive accuracy. XGBoost performed well but with slightly higher error, yielding an R-squared of 0.85 and p-value of 0.063. In contrast, Prophet had the highest forecast errors, with an R-squared of 0.78 and p-value of 0.015. For enhanced interpretability, SHapley Additive exPlanations (SHAP) were applied to RFR, identifying lagged current account balance, GDP per capita, and lagged population as key predictors, along with total population and government net lending/borrowing. SHAP plots provided insights into these features' contributions to GDP predictions. This study highlights RFR's effectiveness in economic forecasting and emphasizes the importance of current and lagged economic indicators.

In actual coal mining operations, hazards like coal and gas outbursts pose significant threats to safety. Extracting coal seam gas through boreholes is the main method of coal mine gas disaster management. Currently, coal seam gas pressure often obtained through direct measurement in underground boreholes requires substantial human and material resources. Therefore, by integrating orthogonal experimental design (OED), random forest regression (RFR) algorithms and Bayesian optimization (BO), a model named OED–BO–RFR was proposed for predicting coal seam gas pressure during pre-drainage processes. The OED–BO–RFR model was tested against RFR, deep neural network (DNN) and support vector regression (SVR) models using a test dataset, revealing its substantial superiority in forecasting coal seam gas pressure during pre-drainage. Furthermore, analyzing the importance of various factors in the OED–BO–RFR model’s construction confirmed its scientific robustness and effectiveness. Subsequently, the OED–BO–RFR, RFR, DNN and SVR models were validated in situ for coal seam gas pre-drainage engineering. The OED–BO–RFR exhibited the best performance among the four models. These results demonstrate that the OED–BO–RFR model can swiftly and accurately predict gas pressure in borehole pre-drained coal seams, providing valuable insights and references for gas disaster management in coal mining enterprises.

ABSTRACTRivers act as the principal channels for transporting terrigenous dissolved organic carbon (DOC) to lakes and reservoirs. Satellite remote sensing-based river monitoring is difficult due to the narrow river form and high spatiotemporal heterogeneity of DOC components. The unique advantages of unmanned aerial vehicles (UAVs) facilitate river DOC concentration monitoring. The DOC concentration in 8 major tributaries (average width: 109.62 m) and shoreside of the Lake Chaohu Basin were retrieved via a hyperspectral UAV. The results showed that (1) the DOC concentration was significantly correlated with the water remote sensing reflectance ([Formula: see text]) of 402, 429-438, 440–451 and 458–462 nm in the blue band (r2: 0.11 to 0.13; p

Accurate mapping of above-ground biomass (AGB) is essential for carbon stock quantification and climate change impact assessment, particularly in mountainous areas. This study applies a random forest (RF) regression model to predict the spatial distribution of AGB in Usho (site A) and Utror (site B) forests located in the northern mountainous region of Pakistan. The predicted maps elucidate AGB variations across these sites, with non-forest areas excluded based on an normalized difference vegetation index (NDVI) threshold value of