Browse Articles
Discover research articles across all indexed journals
A machine learning framework for predicting shear strength properties of rock materials
Abstract The shear strength characteristics of rock materials, specifically internal friction angle and cohesion, are critical parameters for the design of rock structures. Accurate strength prediction can significantly reduce design time and costs while minimizing material waste associated with extensive physical testing. This paper utilizes experimental data from rock samples in the Himalayas to develop a novel machine learning model that combines the improved sparrow search algorithm (ISSA) with Extreme Gradient Boosting (XGBoost), referred to as the ISSA-XGBoost model, for predicting the shear strength characteristics of rock materials. To train and validate the proposed model, a dataset comprising 199 rock measurements and six input variables was employed. The ISSA-XGBoost model was benchmarked against other models, and feature importance analysis was conducted. The results demonstrate that the ISSA-XGBoost model outperforms the alternatives in both training and test datasets, showcasing superior predictive accuracy (R² = 0.982 for cohesion and R² = 0.932 for internal friction angle). Feature importance analysis revealed that uniaxial compressive strength has the greatest influence on cohesion, followed by P-wave velocity, while density exerts the most significant impact on internal friction angle, also followed by P-wave velocity.
Force overestimation during vascular occlusion is triggered by motor system inhibition
A data driven approach to urban area delineation using multi source geospatial data
Abstract This study introduces a data-driven, bottom-up approach to urban delineation, integrating feature engineering with the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm, which represents a significant improvement in precision and methodology compared to traditional approaches that rely on simplistic OpenStreetMap (OSM) road node data aggregations. By employing a broad array of OSM categories and refining data selection through feature engineering, our research significantly enhances the precision and relevance of urban clustering. Using Bavaria, Germany, as a case study, we demonstrate that feature engineering effectively reduces noise and mitigates common DBSCAN clustering pitfalls by filtering out irrelevant and autocorrelated data. The robustness of the proposed method is validated through a comprehensive assessment involving three key elements: (1) a 5% improvement in average accuracy, (2) optimal clustering selections based on entropy values that eliminate the need for prior knowledge, and (3) validation through nighttime light data and Zipf’s law, where a high p-value of 0.99 confirms a good fit, supporting the power law. This study contributes to urban studies by providing a scalable, replicable model that incorporates advanced data processing techniques and multidimensional data sources, supporting improved urban planning and policy-making while effectively delineating urban areas in varied settings.
How blowing the whistle on the Theranos scandal transformed Erika Cheung’s career
Enhanced probabilistic prediction of pavement deterioration using Bayesian neural networks and cuckoo search optimization
Abstract The predictive performance of probabilistic pavement condition deterioration is critical for effective maintenance and rehabilitation decisions. Currently, numerous improved models exist, but few rely on probabilistic models to improve pavement deterioration prediction. Therefore, this study proposed an improved probabilistic model for pavement deterioration prediction based on the coupling of Bayesian neural network (BNN) and cuckoo search (CS) algorithm. The model prediction performance is evaluated against two metrics: determination coefficient (R2) and standard deviation (stability). Finally, based on the data from the pavement management system in Shanxi Province, it was verified that the CS-BNN model outperforms the genetic algorithm-BNN, particle swarm optimization-BNN, and BNN models in terms of the two metrics. Sensitivity analysis further confirms the robustness of the CS-BNN model. The findings indicate that the CS-BNN model provides more reliable predictions with lower uncertainty, aiding road engineers in optimizing maintenance schedules and costs.
A dual branch model for predicting microseismic magnitude time series named DTFNet
Enhanced electrochemical detection of dopamine and uric acid using Au@Ni-MOF and employing 2D structure DFT simulation
Strength and microstructural characteristics of sand soils stabilized with paper sludge Ash-Based geopolymer
Sustainable groundwater management through water quality index and geochemical insights in Valsad India
Author Correction: Partner relationships, hopelessness, and health status strongly predict maternal well-being: an approach using light gradient boosting machine
Flax domestication processes as inferred from genome-wide SNP data
Abstract Flax (Linum usitatissimum L.) is one of the founder crops domesticated for oil and fiber uses in the Near-Eastern Fertile Crescent, but its domestication history remains largely elusive. Genetic inferences so far have expanded our knowledge in several aspects of flax domestication such as the wild progenitor, the first use of domesticated flax, and domestication events. However, little is known about flax domestication processes involving multiple domestication events. This study applied genotyping-by-sequencing to infer flax domestication processes. Ninety-three Linum samples representing four flax domestication groups (oilseed, fiber, winter and capsular dehiscence) and its wild progenitor (or pale flax; L. bienne Mill.) were sequenced. SNP calling identified 16,998 SNPs that were widely distributed across 15 flax chromosomes. Diversity analysis found that pale flax had the largest nucleotide diversity, followed by indehiscent, winter, oilseed and fiber cultivated flax. Pale flax seemed to be under population contraction, while the other four domestication groups were under population expansion after bottleneck. Demographic inferences showed that five Linum groups carried clear genetic signals of multiple mixture events that were associated largely with oilseed flax. Phylogenetic analysis revealed that oilseed, fiber and winter flax formed two separate phylogenetic subclades. One subclade had abundant winter flax, along with some oilseed and fiber flax, mainly originating in the Near East and nearby regions. The other subclade mainly had oilseed and fiber flax originating from Europe and other parts of the world. Dating genetic divergences with an assumption of 10,000 years before present (BP) of flax domestication revealed that oilseed and fiber flax spread to Europe 5800 years BP and domestication for winter hardiness occurred in the Near East 5100 years BP. These findings provide new significant insights into flax domestication processes.
Temperature sensor with adjustable frequency band integrated with antenna and perception
Ultimate bearing capacity of embedded strip footings on rock slopes under inclined loading conditions using adaptive finite element limit analysis
A unified approach for weakly supervised crack detection via affine transformation and pseudo label refinement
Optimal satellite selection using quantum convolutional autoencoder for low-cost GNSS receiver applications
Abstract The increasing reliance on global navigation satellite systems for diverse applications necessitates the development of efficient satellite selection methods to optimize positioning accuracy and system performance. In particular, low-cost global navigation satellite systems receivers face challenges in managing data from multiple visible satellites, often resulting in suboptimal performance due to high geometric dilution of precision values. Effective satellite selection is crucial for improving the accuracy and reliability of positioning solutions in these systems. Quantum computing and machine learning provide promising solutions by using data patterns for complex optimization problems. This work proposes the quantum convolutional autoencoder-based optimal satellite selection method. This new satellite selection method examined the data collected from the receiver located at latitude 16.33° N and longitude 80.62° E, collected on March 10, 2022. The main aim is to enhance the performance of low-cost receivers by minimizing the geometric dilution of precision values and optimizing the tetrahedron volume function. Quantum convolutional autoencoders process the satellite data to balance the navigational solution’s computational burden and the navigational algorithm’s accuracy. The model aims to identify the most optimal satellites for positioning by setting geometric dilution of precision as the cost function. The QCAE-based method achieves a CEP of 1.384 m and SEP of 1.759 m for four selected satellites, compared to 5.937 m and 6.691 m for PSOSSM. For nine satellites, QCAE achieves a CEP of 1.287 m and SEP of 1.713 m, while PSOSSM results in 5.725 m and 6.385 m, respectively. Additionally, QCAE reduces computations by over 64%, requiring 730 multiplications and 713 additions, compared to 2034 multiplications and 2017 additions for all visible satellites. This proposed approach provides the optimal navigation solution for cost-effective implementations in a real-time environment. This research provides new insights into satellite selection strategies using machine learning approaches.