Ensemble machine learning prediction accuracy: local vs. global precision and recall for multiclass grade performance of engineering students by Yagyanath Rimal, Yagyanath Rimal, Navneet Sharma

“…The grid search for random forest algorithms achieved a score of 79% when optimally tuned; however, the training accuracy was 99%. …”
Get full text

Multilayer perceptron deep learning radiomics model based on Gd-BOPTA MRI to identify vessels encapsulating tumor clusters in hepatocellular carcinoma: a multi-center study by Mengting Gu, Wenjie Zou, Huilin Chen, Ruilin He, Xingyu Zhao, Ningyang Jia, Wanmin Liu, Peijun Wang

“…Compared with the two models aforementioned, the Radiology MLP model demonstrated a 33.4%-131.3% improvement in NRI and a 9.3%-50% improvement in IDI, showing better discrimination, calibration and clinical usefulness in three sets, which was selected as the optimal predictive model. …”
Get full text

Research status of gearboxes for high-power nuclear power circulation pumps by WEI Fude, WANG Huanhuan, YANG Shuncheng

“…Many scholars are committed to in-depth research on high-power nuclear power circulating pump gearboxes from the perspectives of dynamics, mechanical design, big data, optimization algorithms, etc., in response to the above-mentioned issues. …”
Get full text

Fuzzy deep learning architecture for cucumber plant disease detection and classification by Anas Bilal, Junaid Ali Khan, Abdulkareem Alzahrani, Khalid Almohammadi, Maha Alamri, Xiaowen Liu

“…At the same time, the ReLU transfer function ensures robustness, mainly when dealing with noisy or incomplete image segments. Feature vector optimization is performed using a chaotic particle swarm algorithm, enhancing the model’s overall accuracy, reliability, and ease of implementation. …”
Get full text

Deep Reinforcement Learning Assisted UAV Path Planning Relying on Cumulative Reward Mode and Region Segmentation by Zhipeng Wang, Soon Xin Ng, Mohammed EI-Hajjar

“…The proposed region segmentation algorithm and cumulative reward model have been tested in different DRL techniques, where we show that the cumulative reward model can improve the training efficiency of deep neural networks by 30.8% and the region segmentation algorithm enables deep Q-network agent to avoid 99% of local optimal traps and assists deep deterministic policy gradient agent to avoid 92% of local optimal traps.…”
Get full text

Mixed reality infrastructure based on deep learning medical image segmentation and 3D visualization for bone tumors using DCU-Net by Kun Wang, Yong Han, Yuguang Ye, Yusi Chen, Daxin Zhu, Yifeng Huang, Ying Huang, Yijie Chen, Jianshe Shi, Bijiao Ding, Jianlong Huang

“…Conclusion: The deep learning DCU-Net algorithm model can improve the performance of tumor CT image segmentation, and the reconstructed fine model can better reflect the actual situation of individual tumors; the MR system constructed based on this model enhances clinicians’ understanding of tumor morphology and spatial relationships. …”
Get full text

Power capacity enhancement of hydropower plant through the penetration of solar and wind energy by Ahmad Shah Irshad, Ahmad Shah Amin, Amir Mohammad Ilham, M.H. Elkholy, Said Elias, Tomonobu Senjyu

“…A multi-objective genetic algorithm was employed to optimize this integration, addressing objectives such as maximizing power output, improving energy efficiency, and minimizing environmental impact. …”
Get full text

Distribution Network Topology Identification Based on Finite Key Nodes and Wasserstein Distance by Yao ZHAO, Yongjiang CHEN, Kunhua JI, Yun WANG

“…Firstly, the finite key nodes can be used to identify the topology when the subspace perturbation model is used to prove the topology change of the distribution network, and the concept of influence degree is introduced through the entropy method based hybrid K-Shell algorithm and the importance of nodes is obtained by the influence degree and the electrical distance between the nodes, thus determining the key nodes in the distribution network topology. …”
Get full text

Facial emotion based smartphone addiction detection and prevention using deep learning and video based learning by C. Joseph, P. Uma Maheswari

“…Additionally, face emotion detection algorithms tuned with MnasNet-Teaching Learning Based Optimization (TLBO) and Convolution Neural Networks (CNN)-Cuckoo Search Optimization (CSO) are employed for accurate emotion recognition. …”
Get full text

Projecting Forest Fire Probability in South Korea Under Climate Change, Population, and Forest Management Scenarios Using AI & Process-Based Hybrid Model (FLAM-Net) by Hyun-Woo Jo, Myoungsoo Won, Florian Kraxner, Seong Woo Jeon, Yowhan Son, Andrey Krasovskiy, Woo-Kyun Lee

“…Enhancements included improving backpropagation for optimization and introducing algorithms for national-specific fire ignition dynamics. …”
Get full text

Long Short-Term Memory–Model Predictive Control Speed Prediction-Based Double Deep Q-Network Energy Management for Hybrid Electric Vehicle to Enhanced Fuel Economy by Haichao Liu, Hongliang Wang, Miao Yu, Yaolin Wang, Yang Luo

“…The initial learning rate and dropout probability of the LSTM speed prediction model are optimized using a Double Deep Q-Network (DDQN) algorithm. …”
Get full text

Spatial pattern and heterogeneity of green view index in mountainous cities: a case study of Yuzhong district, Chongqing, China by Aiping Gou, Xuyuan Wang, Jiangbo Wang, Chenjie Wang, Guanzheng Tan

“…This study investigates the spatial heterogeneity of the GVI and its influencing factors in Yuzhong District, Chongqing, a mountainous city in China. Deep learning algorithms were employed to calculate the green visibility of street view images, and Geographic Weighted Regression (GWR) and the Optimal Parameter-Based Geodetector (OPGD) were utilized to analyze the relationships between GVI and factors such as road physical attributes, the Normalized Difference Vegetation Index (NDVI), and topographic features. …”
Get full text

An integrated AHP-GP-GA approach for public project portfolio selection problem by A.O. Mogbojuri, O.A. Olanrewaju

“…The study presents integrated models of the Analytic Hierarchy Process, Goal Programming, and Genetic Algorithm (AHP-GP-GA) by removing the bias of each model for Public PPSP and building a relationship between the developed models.The AHP method was utilized to establish project selection criteria, allocate relative priority values to stakeholders, and calculate the overall weighting of project alternatives. …”
Get full text

FedEmerge: An Entropy-Guided Federated Learning Method for Sensor Networks and Edge Intelligence by Koffka Khan

“…Empirically, FedEmerge improves global model accuracy and convergence speed on highly skewed non-IID benchmarks, and it reduces performance disparities among clients compared to FedAvg. …”
Get full text

Digital Intelligence Pathology Platform and Its Service Pattern by Xiaohong Chen, Liu Liu, Yajua Niu, Xiaoliang Liu, Xiaohai Li, Jianhua Zhou, Junpu Wang

Get full text

Laser-induced Breakdown Spectroscopy Based on Pre-classification Strategy for Quantitative Analysis of Rock Samples by Weiheng KONG, Lingwei ZENG, Yu RAO, Sha CHEN, Xu WANG, Yanting YANG, Yixiang DUAN, Qingwen FAN

“…Different element quantitative models were constructed for each rock type. The kNN algorithm was selected using cross-validation to determine the optimal k value, and the key punishment parameter C and RBF width parameter γ of the SVM algorithm were determined using a grid search method. …”
Get full text

Integrating data from unmanned aerial vehicles and Sentinel-2 with PROSAIL-5D-driven machine learning for fuel moisture content estimation in agroecosystems by Jinlong Liu, Jia Jin, Jing Huang, Mengjuan Wu, Shaozheng Hao, Haoyi Jia, Tengda Qin, Yuqing Huang, Dan Chen, Nathsuda Pumijumnong

“…A genetic algorithm-optimized backpropagation neural network was then applied to assess the effectiveness of the fused remote sensing data and PROSAIL-5D simulation in improving FMC retrieval accuracy. …”
Get full text

CFD method for the effect of baffle locations and baffle lengths on the hydraulic characteristics of a horizontal sedimentation tank by T. Zhang, W. Wei, Y. Hong, J. Wang, J. Zhang, Y. Pei, Q. Li

“…The governing equations were discretized by using the finite volume method (FVM), the free water surface was tracked by the volume of fluid (VOF) method, and the coupling between velocity and pressure was solved by using the pressure implicit splitting-operator (PISO) algorithm. The results show that, in the design of baffle location and baffle length of a laboratory horizontal sedimentation tank, a baffle with a proper length and arranged at a certain proper distance horizontally away from the inlet of the tank will result in smallest circulation area in the flow field, and can effectively improve its operational efficiency; moreover, changing the length of the baffle cannot obviously reduce the circulation area, that is, the effect of baffle location is greater than that of baffle length on the hydraulic characteristics of a horizontal sedimentation. …”
Get full text

Analytical Solution and Analysis of Aerodynamic Noise Induced by the Turbulent Flow Interaction of a Plate with Double-Wavelength Bionic Serration Leading Edges by Chenye Tian, Xiaomin Liu, Lei Wang, Yuefei Li, Yandong Wu

“…In this study, in order to reduce the aerodynamic noise of a flat plate operating in a steady uniform flow, double-wavelength leading-edge serrations based on Ayton’s analytical model are optimized by the meta-heuristic optimization algorithm. …”
Get full text

Real-time temperature prediction of large-scale lithium battery module driven by data based on few measurement points by Jiajie HAN, Qingyang YUAN, Yu LI, Bo ZHANG, Ke XUE, Tian LAN

“…This slight error increase is attributed to the complex heat transfer dynamics near cooling interfaces, which pose challenges for most data-driven models. This study highlights the effectiveness of the Gappy POD algorithm in managing the thermal dynamics of large-scale energy storage systems in real time. …”
Get full text