Appraisal of numerous machine learning techniques for the prediction of bearing capacity of strip footings subjected to inclined loading

Modeling Earth Systems and Environment, Journal Year: 2024, Volume and Issue: 10(3), P. 4067 - 4088

Published: April 15, 2024

Language: Английский

---

Plant roots reduce rill detachment and shallow instability in forest topsoils

Rhizosphere, Journal Year: 2024, Volume and Issue: 31, P. 100921 - 100921

Published: June 26, 2024

Language: Английский

---

Enhancing Accuracy in Predicting Continuous Values through Regression

Iraqi Journal for Computer Science and Mathematics, Journal Year: 2025, Volume and Issue: 5(4)

Published: Jan. 6, 2025

Enhancing the accuracy in predicting continuous values remains a significant challenge, especially when dealing with imbalanced data and choosing appropriate models. Regression techniques are widely used mining, machine learning fields for this purpose. However, traditional algorithms struggle to achieve high because of limitations complex distribution. This study addresses these gaps by proposing new framework that evaluates multiple regression models using Boston House Pricing Dataset (BHD). The examined involve simple linear, Polynomial, Lasso, Ridge, Random Forest, Keras Gradient Boosting regression. compared evaluation metrics such as R-squared Score (R2), Mean Squared Error (MSE), Absolute (MAE). Among models, first promising outcomes indicate Forest Ridge regressors scored level R2 i.e. 89.9 88.3, respectively. In addition, model offers best result 92 MSE 0.72 MAE 2.00. To further enhance model, research applies two techniques. Re-sampling optimization RandomizedSearchCV tuned hyper-parameter improved score 93.2 better 0.015 0.82. These findings prove improvement performance offer potential practical application real-world scenarios.

Language: Английский

---

Research Trends in Wildland Fire Prediction Amidst Climate Change: A Comprehensive Bibliometric Analysis

Forests, Journal Year: 2024, Volume and Issue: 15(7), P. 1197 - 1197

Published: July 10, 2024

Wildfire prediction plays a vital role in the management and conservation of forest ecosystems. By providing detailed risk assessments, it contributes to reduction fire frequency severity, safeguards resources, supports ecological stability, ensures human safety. This study systematically reviews wildfire literature from 2003 2023, emphasizing research trends collaborative trends. Our findings reveal significant increase activity between 2019 primarily driven by United States Forest Service Chinese Academy Sciences. The majority this was published prominent journals such as International Journal Wildland Fire, Ecology Management, Remote Sensing, Forests. These publications predominantly originate Europe, States, China. Since 2020, there has been substantial growth application machine learning techniques predicting fires, particularly estimating occurrence probabilities, simulating spread, projecting post-fire environmental impacts. Advanced algorithms, including deep ensemble learning, have shown superior accuracy, suggesting promising directions for future research. Additionally, integration with cellular automata markedly improved simulation behavior, enhancing both efficiency precision. profound impact climate change on also necessitates inclusion extensive data predictive models. Beyond conventional studies focusing behavior forecasting consequences fires become integral formulating more effective strategies. concludes that regional disparities knowledge exist, underscoring need capabilities underrepresented areas. Moreover, is an urgent requirement enhance artificial intelligence intensify efforts identifying leveraging various drivers refine accuracy. insights generated field will profoundly augment our understanding prediction, assisting policymakers practitioners managing resources sustainably averting calamities.

Language: Английский

---

Impact of climate change on spatiotemporal patterns of snow hydrology: Conceptual frameworks, machine learning versus nested model

Physics and Chemistry of the Earth Parts A/B/C, Journal Year: 2024, Volume and Issue: 136, P. 103691 - 103691

Published: Aug. 10, 2024

Language: Английский

---

From tree to plot: investigating stem CO₂ efflux and its drivers along a logging gradient in Sabah, Malaysian Borneo

New Phytologist, Journal Year: 2024, Volume and Issue: 244(1), P. 91 - 103

Published: Aug. 15, 2024

Stem respiration constitutes a substantial proportion of autotrophic in forested ecosystems, but its drivers across different spatial scales and land-use gradients remain poorly understood. This study quantifies examines the impact logging disturbance on stem CO

Language: Английский

---

Exploring the accuracy of Random Forest and Multiple Regression models to predict rill detachment in soils under different plant species and soil treatments in deforested lands

Modeling Earth Systems and Environment, Journal Year: 2023, Volume and Issue: 10(2), P. 2533 - 2546

Published: Dec. 28, 2023

Language: Английский

---

Effects of tree and shrub species on soil quality, sediment detachment capacity caused by rills and surface slope stability in forest lands of Northern Iran

International Journal of Sediment Research, Journal Year: 2024, Volume and Issue: 39(5), P. 795 - 803

Published: July 6, 2024

A root system is an important factor to increase soil resistance detachment of particles. However, due the large number species, there a need for studying impacts native plant species on quality and erodibility. This investigation did flume experiments at various slopes (9.2%, 18.1%, 25.1%, 32.5%) different water flow rates (0.56, 0.67, 0.74, 0.81, 0.94 L/(m·s)), evaluate sediment capacity caused by rills (Dc) rill erodibility (Kr) as well hillslopes with three common including Carpinus betulus (as natural tree species), Alnus glutinosa planted species) Mespilus germanica shrub in forestland northern Iran. The variability Dc has been associated properties characteristics betulus. was significantly lower (average, −45%) soils under compared two other (p < 0.01). higher values medium weight diameter aggregates (MWD), organic carbon (OC), total nitrogen (TN), phosphorous (TP), potassium (K), calcium (Ca), magnesium (Mg) more extended system, confirmed negative correlations between studied variables. Kr also among species. played useful role increasing erosion yielding safety (1.61) forest ecosystem. Overall, current study supports broader use (such betulus) areas exposed surface instability, effective eco-engineering conservation technique alternative technology instead utilizing artificial expensive management practices.

Language: Английский

---

A review of optimization and decision models of prescribed burning for wildfire management

Risk Analysis, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 23, 2024

Abstract Prescribed burning is an essential forest management tool that requires strategic planning to effectively address its multidimensional impacts, particularly given the influence of global climate change on fire behavior. Despite inherent complexity in prescribed burns, limited efforts have been made comprehensively identify critical elements necessary for formulating effective models. In this work, we present a systematic review literature optimization and decision models burning, analyzing 471 academic papers published last 25 years. Our study identifies four main types models: spatial‐allocation, spatial‐extent, temporal‐only, spatial–temporal. We observe growing number studies modeling primarily due expansion spatial‐allocation spatial–temporal There also increase as consider more affecting effectiveness. components models, including stakeholders, variables, objectives, influential factors, enhance model practicality. The examines solution techniques, such integer programming spatial allocation, stochastic dynamic probabilistic multiobjective balancing trade‐offs. These techniques' strengths limitations are discussed help researchers adapt methods specific challenges optimization. addition, investigate general assumptions relaxation Lastly, propose future research develop comprehensive incorporating behaviors, stakeholder preferences, long‐term impacts. Enhancing these models' accuracy applicability will enable decision‐makers better manage wildfire treatment outcomes.

Language: Английский

---

Dynamic patterns and potential drivers of river water quality in a coastal city: Insights from a machine-learning-based framework and water management

Journal of Environmental Management, Journal Year: 2024, Volume and Issue: 370, P. 122911 - 122911

Published: Oct. 15, 2024

Language: Английский

---