Legal Challenges and Responses to Artificial Intelligence-Assisted Decision-Making in the International Economic Law System DOI Open Access
Xiaojuan Zhang

Applied Mathematics and Nonlinear Sciences, Год журнала: 2024, Номер 9(1)

Опубликована: Янв. 1, 2024

Abstract Legal judgment prediction is becoming a research hotspot in the legal field as an important artificial intelligence-assisted decision-making tool case management, which able to predict results. In this paper, data from 2018 China Law Research Cup competition gathered, and dataset preprocessed context of international economic law. Then, multi-task model for verdict proposed, training optimization are designed using CNN, RNN, LSTM semantic coding layer. The proposed paper achieves significant improvement 8% 6% accuracy charging task sentence task, respectively. outcome prediction, improved by 14.6% on average compared feature model-based modeling approach.

Язык: Английский

Rapid Evaluation Method to Vertical Bearing Capacity of Pile Group Foundation Based on Machine Learning DOI Creative Commons
Yanmei Cao, Jing Ni, Jianguo Chen

и другие.

Sensors, Год журнала: 2025, Номер 25(4), С. 1214 - 1214

Опубликована: Фев. 17, 2025

With the continuous increase in bridge lifespans, rapid check and evaluation of vertical bearing capacity for pile foundations existing bridges have been greater demand. The usual practice is to carry out compression tests under static loads order obtain accurate ratio dynamic stiffness. However, it difficult costly conduct situ experiments each foundation. Herein, a method measure proposed. Firstly, 3D-bearing cap-pile group-soil interaction model was established simulate test foundation that subject loads, then numerical results were validated by loading on an abandoned pier with same group foundation; dataset machine learning constructed using results, finally, could be predicted rapidly. show following outcomes: can effectively foundations; intelligent prediction based predict stiffness thus rapidly evaluate residual designed ultimate capacity, allowing nondestructive testing bridges.

Язык: Английский

Процитировано

0

Load-deformation prediction of bored piles using sequential soil profile encoding with transformer architecture: A study of Bangkok subsoil DOI
Sompote Youwai,

Chissanupong Thongnoo

Expert Systems with Applications, Год журнала: 2025, Номер unknown, С. 127085 - 127085

Опубликована: Фев. 1, 2025

Язык: Английский

Процитировано

0

Borehole Breakout Prediction Based on Multi-Output Machine Learning Models Using the Walrus Optimization Algorithm DOI Creative Commons
Rui Zhang, Jian Zhou, Ming Tao

и другие.

Applied Sciences, Год журнала: 2024, Номер 14(14), С. 6164 - 6164

Опубликована: Июль 15, 2024

Borehole breakouts significantly influence drilling operations’ efficiency and economics. Accurate evaluation of breakout size (angle depth) can enhance strategies hold potential for in situ stress magnitude inversion. In this study, borehole is approached as a complex nonlinear problem with multiple inputs outputs. Three hybrid multi-output models, integrating commonly used machine learning algorithms (artificial neural networks ANN, random forests RF, Boost) the Walrus optimization algorithm (WAOA) techniques, are developed. Input features determined through literature research (friction angle, cohesion, rock modulus, Poisson’s ratio, mud pressure, radius, stress), 501 related datasets collected to construct dataset. Model performance assessed using Pearson Correlation Coefficient (R2), Mean Absolute Error (MAE), Variance Accounted For (VAF), Root Squared (RMSE). Results indicate that WAOA-ANN exhibits excellent stable prediction performance, particularly on test set, outperforming single-output ANN model. Additionally, SHAP sensitivity analysis conducted model reveals maximum horizontal principal (σH) most influential parameter predicting both angle depth breakout. Combining results studies analyses conducted, considered be an effective size.

Язык: Английский

Процитировано

3

Enhancing Sustainability of Building Foundations with Efficient Open-End Pile Optimization DOI Open Access
Primož Jelušič

Sustainability, Год журнала: 2024, Номер 16(16), С. 6880 - 6880

Опубликована: Авг. 10, 2024

Optimizing open-end piles is crucial for sustainability as it minimizes material consumption and reduces environmental impact. By improving construction efficiency, less steel needed, reducing the carbon footprint associated with production transportation. Improved pile performance also results in more durable structures that require frequent replacement maintenance, which turn saves resources energy. This paper presents a parametric study on optimal designs open-ended sand, presenting novel approach to directly compute using CPT results. It addresses challenges posed by soil variability layered conditions, optimization model accounting interdependencies among length, diameter, wall thickness properties, including pile–soil plug system. A mixed-integer OPEN-Pile was developed, consisting of an objective function mass CO2 emissions. The constrained set design geotechnical conditions corresponded current codes practice recommendations. efficiency developed illustrated two case studies. In Blessington calculation show economical environmentally friendly increase diameter than length. efficient design, ratio between calculated at upper limit. For optimum ratios length are 5, 50 250, respectively. profile, decision where place base depends resistance cone tip individual layers. To determine layer should be placed, we need perform given data.

Язык: Английский

Процитировано

0

Legal Challenges and Responses to Artificial Intelligence-Assisted Decision-Making in the International Economic Law System DOI Open Access
Xiaojuan Zhang

Applied Mathematics and Nonlinear Sciences, Год журнала: 2024, Номер 9(1)

Опубликована: Янв. 1, 2024

Abstract Legal judgment prediction is becoming a research hotspot in the legal field as an important artificial intelligence-assisted decision-making tool case management, which able to predict results. In this paper, data from 2018 China Law Research Cup competition gathered, and dataset preprocessed context of international economic law. Then, multi-task model for verdict proposed, training optimization are designed using CNN, RNN, LSTM semantic coding layer. The proposed paper achieves significant improvement 8% 6% accuracy charging task sentence task, respectively. outcome prediction, improved by 14.6% on average compared feature model-based modeling approach.

Язык: Английский

Процитировано

0