Random Regression Research Articles

Forest canopy height modelling based on photogrammetric data and machine learning methods

AbstractForest topographic survey is a problem that photogrammetry has not solved for a long time. Forest canopy height is a crucial forest biophysical parameter which is used to derive essential information about forest ecosystems. In order to construct a canopy height model in forest areas, this study extracts spectral feature factors from digital orthophoto map and geometric feature factors from digital surface model, which are generated through aerial photogrammetry and LiDAR (light detection and ranging). The maximum information coefficient, Pearson, Kendall, Spearman correlation coefficients, and a new proposed index of relative importance are employed to assess the correlation between each feature factor and forest vertical heights. Gradient boosting decision tree regression is introduced and utilised to construct a canopy height model, which enables the prediction of unknown canopy height in forest areas. Two additional machine learning techniques, namely random forest regression and support vector machine regression, are employed to construct canopy height model for comparative analysis. The data sets from two study areas have been processed for model training and prediction, yielding encouraging experimental results that demonstrate the potential of canopy height model to achieve prediction accuracies of 0.3 m in forested areas with 50% vegetation coverage and 0.8 m in areas with 99% vegetation coverage, even when only a mere 10% of the available data sets are selected as model training data. The above approaches present techniques for modelling canopy height in forested areas with varying conditions, which have been shown to be both feasible and reliable.

Starting then stopping: a nationwide register-based study on the magnitude, predictors, and urban-rural patterns of under-vaccination variation across health centers in The Gambia

ABSTRACT Objectives Six million children were under-vaccinated in 2022. Our study aimed to 1) quantify the magnitude of under-vaccination variation between health facilities, 2) assess to which extent individual and health center level factors contributed to the variation, 3) identify individual and health facility factors associated with under-vaccination, and 4), explore rural vs. urban health facility variations. Methods We used data from 61,839 children from The Gambia national routine vaccination register. We cross tabulated under-vaccination status across study variables and fitted two-level random intercept multilevel logistic regression models to measure variance, contribution to the variance, and factors associated with the variance and under-vaccination. Results We found that 7% of the prevalence of under-vaccination was due to variation between health facilities. Thirty-seven percent of the variation was explained by individual and health center variables. The variables explained 36% of the variance in urban and 19% in rural areas. Children who were not vaccinated at 4 months or with delayed history, due for vaccination in the rainy season, and health facilities with very small or large population to health worker ratios had higher under-vaccination odds. Conclusion Our study indicates that one of the pathways to improving vaccination coverage is addressing factors driving under-vaccination inequities between health facilities through urban-rural differentiated strategies.

Novel Insights in Soil Mechanics: Integrating Experimental Investigation with Machine Learning for Unconfined Compression Parameter Prediction of Expansive Soil

This paper presents a novel application of machine learning models to clarify the intricate behaviors of expansive soils, focusing on the impact of sand content, saturation level, and dry density. Departing from conventional methods, this research utilizes a data-centric approach, employing a suite of sophisticated machine learning models to predict soil properties with remarkable precision. The inclusion of a 30% sand mixture is identified as a critical threshold for optimizing soil strength and stiffness, a finding that underscores the transformative potential of sand amendment in soil engineering. In a significant advancement, the study benchmarks the predictive power of several models including extreme gradient boosting (XGBoost), gradient boosting regression (GBR), random forest regression (RFR), decision tree regression (DTR), support vector regression (SVR), symbolic regression (SR), and artificial neural networks (ANNs and proposed ANN-GMDH). Symbolic regression equations have been developed to predict the elasticity modulus and unconfined compressive strength of the investigated expansive soil. Despite the complex behaviors of expansive soil, the trained models allow for optimally predicting the values of unconfined compressive parameters. As a result, this paper provides for the first time a reliable and simply applicable approach for estimating the unconfined compressive parameters of expansive soils. The proposed ANN-GMDH model emerges as the pre-eminent model, demonstrating exceptional accuracy with the best metrics. These results not only highlight the ANN’s superior performance but also mark this study as a groundbreaking endeavor in the application of machine learning to soil behavior prediction, setting a new benchmark in the field.

The minimal important difference in orthodontic treatment duration: a survey across adult patients.

The minimal important difference (MID) is defined as the smallest difference that the patient perceives as important. Furthermore, the smallest worthwhile effect (SWE) is the important change measured with the benefit-harm trade-off method. The aim of this study was to evaluate the MID in orthodontic treatment duration to inform the decision regarding seeking procedures to accelerate orthodontic tooth movement and reduce treatment duration. We constructed a survey eliciting views of the MID from adult participants from four countries undergoing orthodontic treatment. Ten questions addressed reduction in the treatment duration for both durations 12 and 24 months, and four questions were related to the reduction in treatment duration that the patients would require to undergo surgical or non-surgical adjunctive procedures. We applied a univariable random effects logistic regression model to examine the association between the participants' characteristics and the MID. Then, we fitted a multivariable logistic random effects regression including significant predictors. Four hundred and fifty adults, with a median age of 21 (interquartile range: 19-24), undergoing orthodontic treatment participated in the survey. Of the respondents, 60% considered 15 days as a trivial reduction from 12 months duration of therapy and 70% considered15 daysa trivial reduction from 24 months. Of the respondents, 48% considered the period of 2 months a moderate reduction from 12 months, and 60% considered 2 months a moderate reduction from 24 months. From these results, we inferred that patients considered reductions of approximately 1 month as the MID in the treatment duration for both 12 and 24 months. However, SWE was considerably more than the MID for most of the participants to decide undergoing surgical adjunctive procedures to reduce the time of therapy. The participants required smaller SWE to undergo non-surgical procedurescompared to surgical procedures. The MID in the treatment duration is one month for both treatment durations 12 and 24 months. Patients require a greater SWE than the MID to undergo adjunctive procedures to shorten the duration, particularly for surgical procedures.

Rock Mineral Volume Inversion Using Statistical and Machine Learning Algorithms for Enhanced Geothermal Systems in Upper Rhine Graben, Eastern France

AbstractAccurately determining the mineralogical composition of rocks is essential for precise assessments of key petrophysical properties like effective porosity, water saturation, clay volume, and permeability. Mineral volume inversion is particularly critical in geological contexts characterized by heterogeneity, such as in the Upper Rhine Graben (URG), where both carbonate and siliciclastic formations are prevalent. The estimation of mineral volumes poses challenges that involve both linear and nonlinear relationships associated with geophysical data. To address this complexity, our methodology strategically integrates the robust insights from standard statistical approaches with three machine learning (ML) algorithms: multi‐layer perceptron, random forest regression, and gradient boosting regression. Furthermore, we propose a new hybrid ensemble model that incorporates a weighted average of multiple ML approaches to predict mineral composition within the Muschelkalk and Buntsandstein formations of the URG. ML techniques for mineral composition prediction in these formations exhibit robust predictive performance. The predicted mineral volumes align closely with quantitative estimates derived from X‐ray diffraction analysis. Additionally, they are in good qualitative agreement with mineral descriptions obtained from cores and cuttings of the Muschelkalk and Buntsandstein formations.

Weekly Versus Bolus Cisplatin Concurrent With Definitive Radiation Therapy for Squamous Carcinoma of the Head and Neck: A Systematic Review and Network Meta-Analysis

The schedule of cisplatin concurrent with definitive radiation for squamous carcinoma of the head and neck remains controversial. Most institutions deliver either a high-dose "bolus" schedule once every 3 weeks or a low-dose weekly schedule. We compared these 2 schedules via a simplified network meta-analysis with a common comparator. We performed a PRISMA-concordant systematic review to identify randomized controlled trials comparing cisplatin with cetuximab for nonmetastatic, locoregionally advanced squamous carcinoma of the head and neck treated with definitive radiation. Trials incorporating primary surgery or induction therapy were excluded. Patient survival times were extracted on a per-event basis from the published curves using a digitizer and validated against published point estimates and hazard ratios (HRs). Survival was compared using random effects Cox regression under a frequentist framework. Toxicity and secondary endpoints were analyzed qualitatively. The Cochrane method assessed the risk of bias. The analysis plan was preregistered with the Open Science Foundation. Five randomized trials were identified, including 1678 patients. There was no statistical difference in overall survival between weekly and bolus regimens (HR, 0.90; 95% CI, 0.53-1.52, P = .345). This Cox model suggested that for the average patient in the cohort, the absolute difference in 5-year overall survival between weekly and bolus regimens was +1.2% (95% CI, -6.1%-+5.9%, P = .345). Secondary endpoints and toxicity were not obviously different by regimen, qualitatively. The cetuximab trials provide indirect data suggesting that the differences between cisplatin schedules are subtle.

Disentangling the impact of nested sources of variability on species growth processes: A mixture of multilevel mixed model approach.

SummaryThe understanding of tree growth processes is crucial for promoting sustainable forest management strategies. This is a challenging task in highly biodiverse ecosystems where many tree species are observed on very few individuals and the small sample sizes hinder a good fit of species‐specific models. We propose the use of finite mixture of random coefficient regression models with multilevel nested random effects to infer guild specific fixed and random effects while evaluating the relative importance of the nested sources of variability on goodness‐of‐fit. This approach extends finite mixture of linear mixed model used for longitudinal or single group structured data contexts. A dedicated expectation–maximisation algorithm is introduced for parameter estimation. Simulations are performed for the evaluation of the misspecification of nested‐grouping structures. This work has been motivated by data collected biennially in Central African rainforests from 1986 to 2010. We show the accuracy of the proposed approach in successfully reproducing individual growth processes and classifying tree species into well‐differentiated clusters with clear ecological interpretations. Moreover, results confirm that interindividual variability appears as the most important factor to explain tropical tree species growth process variability from Central Africa forests.

Exploring the prognostic impact of differences in treatment strategies for SCLC with different histologies and prognostic factors for C-SCLC: A SEER population-based study

BackgroundCombined small cell lung cancer (C-SCLC) is a rare type of small cell lung cancer (SCLC), and it is controversial whether to choose the same treatment regimen as SCLC due to its multiple histologic components. Study methods and resultsRecords of patients with small cell lung cancer diagnosed between 2010 and 2020 were extracted using the SEER database. The OS of patients with different histological types under the same staging and treatment regimen was analyzed. It was found that early-stage (stage IA-IIA) surgical treatment, systemic chemotherapy alone, and chemoradiotherapy were more efficacious than C-SCLC and P-SCLC in patients with limited-stage (P = 0.054, P = 0.001, P = 0.019). In patients with extensive staging, the OS of patients with systemic chemotherapy regimens differed (P = 0.045) and was better in C-SCLC than in P-SCLC. We further explored the treatment strategy for patients with C-SCLC, which was shown by a COX regression model based on prognostic factors screened by Random Forest and LASSO regression models. Surgery, radiotherapy, and chemotherapy would be beneficial for survival. In a subgroup analysis based on stage and treatment regimen, it was shown that patients with early staging (stage IA-IIA) had a better prognosis with surgery (P < 0.001); in patients with extensive staging, chemoradiotherapy was favorable to the patient's prognosis (P = 0.022). ConclusionBoth limited-stage and extensive-stage C-SCLC patients are more sensitive to chemotherapy than P-SCLC patients. Patients with C-SCLC who have access to surgery should undergo surgery as early as possible, while chemoradiotherapy is recommended for patients with extensive staging. Patient age, gender, tumor size, surgery, chemotherapy, radiotherapy, and metastasis may individually affect patient prognosis.

Soil moisture estimation using ground scatterometer and Sentinel-1 data

Soil moisture (SM) is a crucial criterion for agronomics and the management of water resources, particularly in areas where the socio-economic status and significant source of income depend upon agriculture and related sectors. This paper intends to estimate SM over the vegetative area using a generalized regression neural network (GRNN) and ground scatterometer and compare the results with SM retrieved using Sentinel-1 data. At the same time, random forest regression (RFR) and support vector regression (SVR) models are used for SM estimation. Correlation analysis results concluded that L-band HV-polarization at 300 incidence angle showed the highest correlation with the measured field parameters. This study investigated backscattering coefficients, VV/VH polarization ratio and polarization phase difference over wheat’s entire growth phase to estimate SM. The results indicate that the GRNN with backscattering coefficients and polarization ratio provided the highest accuracy compared to the random forest (RF) and SVR with the root mean square error of 0.093 over the Yavatmal District, Maharashtra, India.

Comparing the allometric model to machine learning algorithms for aboveground biomass estimation in tropical forests

Estimating aboveground biomass (AGB) pools is critical for understanding the structural characteristics of forest ecosystems, as the quantitative analysis of biomass contributes to a better understanding of carbon sequestration and in studying the productivity within forest ecosystems. A more precise estimation of AGB is necessary for the estimation of the amount of carbon stored in an ecosystem that is needed for the valuation of the ecosystem services. Though regression models have been used previously for both tree-level and stand-level biomass estimations, supervised machine learning models have rarely been put to test in field-based estimations of AGB. This study was done in the tropical forests of Northeast India (two reserve forests in Tripura) to compare the allometric performance of weighted nonlinear regression model (NLM) to random forest regression (RFR), extreme gradient boost regression (XGB), and support vector machine regression (SVR) in estimating tree-level AGB. Due to their ease of measurement in field conditions and ensuring practical use, tree diameter at breast height (DBH) and tree height (H) were used as the independent variables to train the models. 506 trees representing 21 local tree species were inventoried. Then, the four models (NLM, RFR, XGB, and SVM) were trained separately on the data and compared to find an optimal method to estimate AGB using modeling efficiency of fit (MEfit) and Root Mean Squared Error (RMSE). The models were then validated using the predicted residual error sum of squares (PRESS) statistics. The best fit model in our study was using RFR, with a high MEfit (0.95–0.99), low RMSE (63.10–132.39 kg), and outperforming all other methods in the PRESS statistics, with the highest modeling efficiency of prediction (0.94–0.99) and lowest mean of absolute PRESS residuals (34.99–78.23). The results from Northeast India presented here provide preliminary evidence that RFR could be a reliable method for determining the field-level carbon stock and understanding the biomass dynamics for evergreen and moist deciduous tropical forests. The model accuracy may be constrained by a limited number of variables chosen to suit practical use in the field, and incorporating additional tree-level or stand-level variables such as crown diameter and stand basal area could enhance the robustness and generalizability of AGB estimation.

Exploring CrossFit performance prediction and analysis via extensive data and machine learning.

The analysis of athletic performance has always aroused great interest from sport scientist. This study utilized machine learning methods to build predictive models using a comprehensive CrossFit (CF) dataset, aiming to reveal valuable insights into the factors influencing performance and emerging trends. Random forest (RF) and multiple linear regression (MLR) were employed to predict performance in four key weightlifting exercises within CF: clean and jerk, snatch, back squat, and deadlift. Performance was evaluated using R-squared (R2) values and mean squared error (MSE). Feature importance analysis was conducted using RF, XGBoost, and AdaBoost models. The RF model excelled in deadlift performance prediction (R2=0.80), while the MLR model demonstrated remarkable accuracy in clean and jerk (R2=0.93). Across exercises, clean and jerk consistently emerged as a crucial predictor. The feature importance analysis revealed intricate relationships among exercises, with gender significantly impacting deadlift performance. This research advances our understanding of performance prediction in CF through machine learning techniques. It provides actionable insights for practitioners, optimize performance, and demonstrates the potential for future advancements in data-driven sports analytics.

Recovering 3D Basin Basement Relief Using High‐Precision Magnetic Data Through Random Forest Regression Algorithm: A Case Study of Tianzhen‐Yanggao Sag in Datong Basin

AbstractInversion of magnetic basement interfaces in basins is essential for interpreting potential field data and studying geothermal resource distribution, as well as basin formation and evolution. This paper introduces a novel method for inverting magnetic basement interfaces using a random forest regression (RFR) algorithm that combines potential field processing and machine learning techniques. The method creates magnetic base interface models and corresponding magnetic anomaly data through the random midpoint displacement method and magnetic interface finite element forward simulation. These anomalies are then processed using techniques such as directional transformations, analytical continuation, spatial derivatives, and fractional transformations. Feature attributes are extracted, and Gini importance is utilized to measure the contributions of feature factors, identify effective features, and enhance algorithm efficiency. The validity and practicality of the method are demonstrated through the analysis of both idealized and noisy models. The proposed machine learning‐based approach is more intelligent, efficient, and accurately represents the relief of magnetic base interfaces. When applied to magnetic survey data in the Datong Basin, it produced a reliable basin base model that aligns with known structural information, paving the way for further research in magnetic interface inversion.

Reducing the risk of grade 3 or greater radiation pneumonitis after radical radiotherapy to the lung.

e20008 Background: Radiation pneumonitis (RP) is the most common adverse response in patients with lung cancer receiving thoracic radiotherapy. Higher grade RP is more likely to lead to mortality and poor quality of life, which could be abated by rigorous treatment standards formulated with individualized clinical characteristics. In this study, we aimed to identify the best clinical prognostic model for evaluating patients with lung cancer after radical radiotherapy. Methods: We collected information of RP patients in recent years and established the new consensus on the diagnosis and treatment of radiation-related pneumonitis. In a nutshell, clinical data of patients treated with radical radiotherapy were collected from August 2020 to August 2022, and part of patients from August 2021 to August 2022 was been intervented by treatment methods. The interventions included limited irradiation of the planning target volume (PTV), individualized lung dose limitation, and standardized steroid therapy. Clinical characteristics including baseline and treatment data were obtained from 693 patients, including 623 patients in the training cohort and 70 patients in the test cohort. Three models were built using different screening methods, including multivariate logistics regression (MLR), backward stepwise regression (BSR), and random forest regression (RFR), to evaluate their prognostic power. Overoptimism in the training cohorts was evaluated by four validation methods including hold-out, 10-fold, leave-one-out, and bootstrap methods, and extra data were used to evaluate the predictive performance of the model. Model calibration, Decision curve analysis (DCA), and evaluation of the nomograms for the three models were completed. Results: The incidence of RP was significantly decreased after the interventions compared to before (68.80% vs. 59.56%, P &lt; 0.05), and the probability of grade 3 or higher RP also decreased from 11.97% to 6.67% (P &lt; 0.05). A model of intervention, interstitial lung disease (ILD), concurrent chemoradiotherapy, standardized steroids, carbon monoxide diffusing capacity (DLCO) &gt; 86.9%, and total lung volume exceeding 5 Gy (V5) &gt; 35.2% for the radiological parameter had the best discriminative ability with an area under the curve of 0.963 (95% CI: 0.938–0.989) in the training chort . The calibration curve showed good agreement between the predicted and actual values, and the DCA showed a positive net benefit for the final model based on the nomogram. Conclusions: The implementation of standardized interventions for the prevention and treatment of RP accompanied with standardized use of steroids resulted in a significant decrease in the incidence of grade 3 or higher RP. Early intervention and methods for reducing the complications of high-dose steroids will be explored in future work.

Association between wildfire PM2.5 and genitourinary cancer survival in California.

10567 Background: Aerosolized particulate matter (PM) 2.5 (i.e., particles ≤ 2.5 µm in diameter or fine particles), including those that emanate from wildfires, has been previously associated with cancer risk and survival. While prior studies have focused primarily on etiology, little is known about the effects of wildfire PM2.5 on cancer prognosis. We sought to explore the effects of wildfire PM2.5 on genitourinary (GU) cancer survival in the state of California where episodic wildfires are considered a component of its climate. Methods: Using the California Cancer Registry, we identified incident prostate (n=81,261), urothelial (n=26,689), and kidney (n=26,303) cancers diagnosed between 2017-2020. Daily PM2.5 was estimated over the same period. Random forests regression was used to fuse diverse datasets (official air quality monitoring, satellite observations, meteorological modeling, predictive smoke modeling, and low-cost sensor networks [PurpleAir]) and produce a consensus estimate of PM2.5 for each centroid of a 1-square kilometer grid across California. We used home address at the time of cancer diagnosis to map each cancer case to the nearest centroid, and averaged PM2.5 exposures over the 6 months preceding cancer diagnosis for each case. Cox-proportional hazards regression was used to quantify the hazard ratio (HR) for death from any cause following cancer diagnosis associated with PM2.5, adjusting for key covariates including age, sex, race/ethnicity, neighborhood socioeconomic status, comorbidity, smoking status, rural/urban residence, and stage at diagnosis. Results: In patients with prostate cancer, a 10 µg/m³ increase in PM2.5 concentration was associated with an 18% increased hazard of death (HR=1.18; 95% confidence interval (CI)= 1.02-1.36, p=0.023). When stratified by stage at diagnosis, the association of PM2.5 and survival was limited to stages I, II, and III (HR=1.39; CI= 1.11-1.74, p=0.005) versus stage IV (HR=1.06; CI= 0.87-1.27, p=0.566). For urothelial cancer, a 10 µg/m³ increase in PM2.5 concentration was associated with a non-statistically significant 11% increased hazard of death (HR=1.11; CI= 0.96-1.29, p=0.151), which did not differ by stage at diagnosis. Kidney cancer survival was not associated with PM2.5 (HR= 0.94; CI= 0.79-1.11, p= 0.44). Conclusions: These findings suggest that higher PM2.5 exposure could influence prostate and urothelial cancer prognosis, perhaps through alteration of physiological pathways relevant to cancer aggressiveness or impairing response to treatment, including cellular immune response, inflammation, angiogenesis, or cell-signaling pathways. Further studies should look into the mechanisms and design mitigating strategies in patients diagnosed with these high-risk cancers who have been exposed to wildfire PM2.5.

Satellites Reveal Spatial Heterogeneity in Dryland Photovoltaic Plants' Effects on Vegetation Dynamics

AbstractLarge‐scale photovoltaic (PV) plants are growing rapidly in drylands because of the rich solar radiation and vast unutilized land. The transformation of landscapes in dryland has threatened local fragile vegetation. Existing studies have investigated the issue by field observations and satellite data, yet the spatial differences in vegetation changes due to dryland PV plants deployment and underlying driving mechanisms remain poorly understood. In this study, Landsat Normalized Difference Vegetation Index data were used to assess the vegetation changes disturbed by PV plants in China's drylands. We further identified environmental factors affecting vegetation changes based on random forest regression model. Our findings reveal the spatial heterogeneity in the impact of PV plants on vegetation dynamics—PV plant deployment promoted the growth of vegetation in the vast majority of arid and hyper‐arid drylands, while it tends to cause vegetation decline in the sub‐humid and semi‐arid drylands. The impact of PV plants on vegetation dynamics depends on local environmental conditions. We found that deploying PV plants in areas with sparse vegetation, low humidity, and long sunshine duration is more likely to promote vegetation restoration. The findings and data maps with highly detailed information can help guide solar energy operators in siting and ecological restoration to enhance techno–ecological synergies in the future.

Climatically robust multiscale species distribution models to support pronghorn recovery in California.

We combined two climate-based distribution models with three finer-scale suitability models to identify habitat for pronghorn recovery in California now and into the future. We used a consensus approach to identify areas of suitable climate now and future for pronghorn in California. We compared the results of climate models from two separate hypotheses about their historical ecology in the state. Under the migration hypothesis, pronghorn were expected to be limited climatically by extreme cold in winter and extreme heat in summer; under the niche reduction hypothesis, historical pronghorn of distribution would have better represented the climatic limitations of the species. We combined occurrences from GPS collars distributed across three populations of pronghorn in the state to create three distinct habitat suitability models: (1) an ensemble model using random forests, Maxent, classification and regression Trees, and a generalized linear model; (2) a step selection function; and (3) an expert-driven model. We evaluated consensus among both the climate models and the suitability models to prioritize areas for, and evaluate the prospects of, pronghorn recovery. Climate suitability for pronghorn in the future depends heavily on model assumptions. Under the migration hypothesis, our model predicted that there will be no suitable climate in California in the future. Under the niche reduction hypothesis, by contrast, suitable climate will expand. Habitat suitability also depended on the methods used, but areas of consensus among all three models exist in large patches throughout the state. Identifying habitat for a species which has undergone extreme range collapse, and which has very fine scale habitat needs, presents novel challenges for spatial ecologists. Our multimethod, multihypothesis approach can allow habitat modelers to identify areas of consensus and, perhaps more importantly, fill critical knowledge gaps that could resolve disagreements among the models. For pronghorn, a better understanding of their upper thermal tolerances and whether historical populations migrated will be crucial to their potential recovery in California and throughout the arid Southwest.

Finding Earthquake Patterns with Time Series and Regression Models

Due to their unpredictability, earthquakes are certainly the most dangerous natural disasters. Scientists have long struggled to understand the movement of tectonic plates, as they occur at depths that cannot be directly seen and occur extremely slowly, at a rate that results in possibly one every thousand years between two adjacent plates. What if we could take a different approach, however? What if we could leverage the recent advances in machine learning to close this gap? That is exactly what we aimed to do in our research, with the question of whether various factors related to earthquakes, such as their magnitude and time of occurrence, could be determined through the analysis of years of historical data. This was conducted through the analysis of a dataset which contained records of all major earthquakes (magnitude 5.5 or higher) from 1900-2023. The library scikit-learn was used for various regression models, such as linear regression, random forest regression, and lasso regression, and the library FB prophet was used for time series forecasting, a method used to analyze patterns in previous earthquakes to predict new ones. Our conclusions were that, with some degree of error, the magnitude of an earthquake can be determined based on location/time and the time of occurrence can be pinpointed to specific days of the week as well as months.

DelayDetect: Pre-takeoff Delay Anticipation

Abstract: Flight delays pose significant challenges to both passengers and airlines, leading to inconvenience, financial losses, and operational disruptions. A comprehensive system is introduced to address the issue of airline flight delays through the application of machine learning models. Leveraging comprehensive datasets encompassing historical flight records, meteorological data, airport congestion patterns and other variables, this model aims to construct predictive models capable of accurately forecasting the probability of flight delays. Various machine learning methodologies including random forests and regression models, are systematically examined in this predictive task. Performance evaluation is conducted employing established metrics such as accuracy and precision. In conclusion, flight delay anticipation is a valuable tool that can revolutionize the aviation industry by minimizing disruptions, reducing costs, and increasing passenger satisfaction.

Optimizing a machine learning design of dielectric properties in lead-free piezoelectric ceramics

Designing lead-free piezoelectric ceramics with tailored electrical properties remains a critical challenge for various applications. In this paper we present a novel methodology integrating Machine Learning (ML) and optimization procedures to fine-tune electrical properties in lead-free (1-x) Na0.5 Bi0.5 TiO3 - x CaTiO3 piezoelectric ceramics. A comprehensive dataset of dielectric measurements serves as the foundation for training ML models that accurately predict the permittivity (ε′) and dielectric loss (tan δ) as functions of Ca2+ concentration (x % Ca), temperature and frequency. Two ML techniques are evaluated: random forest regression, and Multi-Layer Perceptron neural network Regression (MLPR). The MLPR model exhibited a superior regression performance, achieving a correlation coefficient of 0.931 and a root mean squared error of 0.029. The MLPR was then optimized by the Non-dominated Sorting Genetic Algorithm II (NSGA-II) to maximizes ε′ while minimizes tan δ. Within the NSGA-II framework, the optimal values were found at the Pareto curve knee, corresponding to a frequency, temperature, and x % Ca of 609.739 kHz, 398.15 K, and 6.10, respectively, resulting in ε′ equal to 857.87 and tan δ equal to 0.0120. This approach demonstrates the effectiveness of combining ML and optimization for designing the electrical properties of piezoelectric ceramics, paving the way for more efficient and targeted material development.

Artificial Intelligence Models for Predicting Ground Vibrations in Deep Underground Mines to Ensure the Safety of Their Surroundings

Ground vibrations induced by underground mining blasting has a significant impact on the stability and safety of surface buildings near mines. Due to the thick rock layers overlying underground mines, there is presently limited accuracy in regard to predicting ground vibrations induced by underground mine blasting. Therefore, this study aims to improve the accuracy of predicting ground vibrations induced by underground blasting by comprehensively measuring the peak particle velocity (PPV) in all three directions and independently considering on the impact of vertical distance. Random forest regression (RFR), bagging regression (BR), and gradient boosting regression (GBR) were used to regress the X-axis PPV (X-PPV), Y-axis PPV (Y-PPV), and Z-axis PPV (Z-PPV) based on blasting records measured at an iron mine. In addition, a genetic algorithm, gray wolf optimizer (GWO), and a particle swarm optimization were used to optimize the parameters of the RFR, BR, and GBR. The comparison results show that GWO-GBR is the optimal model for the prediction of the X-PPV (R2 = 0.8072), Y-PPV (R2 = 0.9147), and Z-PPV (R2 = 0.9265), respectively. Thus, the GWO-GBR model proposed in this study is considered a highly reliable model for predicting ground vibrations induced by underground mine blasting to ensure the safety of the mines’ surroundings.