Article
Computer Science, Artificial Intelligence
Vasileios Christou, Kyriakos Koritsoglou, Georgios Ntritsos, Georgios Tsoumanis, Markos G. Tsipouras, Nikolaos Giannakeas, Evripidis Glavas, Alexandros T. Tzallas
Summary: This paper presents a system that utilizes a hybrid machine learning algorithm to improve the accuracy of a low-cost temperature sensor. The system collects data from a series of sensors using a low-cost single-board computer and uses the proposed linear regression heterogeneous hybrid extreme learning machine algorithm to increase the sensor's accuracy.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Construction & Building Technology
Hui Li, Weizhong Chen, Xuyan Tan, Xianjun Tan
Summary: This paper proposes a novel back analysis method, which combines the optimized particle swarm optimization (OPSO) algorithm with the support vector machine (SVM) algorithm. The method, named OPSO-SVM-ABAQUS, is further integrated with the finite element method (FEM). The proposed algorithm accurately estimates the geomechanical parameters by analyzing the monitoring displacement data, and it is compared with six other algorithms, demonstrating its feasibility in complex geological conditions.
TUNNELLING AND UNDERGROUND SPACE TECHNOLOGY
(2023)
Article
Multidisciplinary Sciences
Hanghang Yan, Kaiyun Liu, Chong Xu, Wenbo Zheng
Summary: This paper introduces the Gaussian process regression (GPR) algorithm to make up for the shortcomings of existing intelligent inversion methods. An improved Gaussian process regression (IGPR) algorithm is proposed by adding two single kernel functions. The particle swarm optimization (PSO) is combined with the IGPR model to optimize the parameters. The application case shows that the PSO-IGPR hybrid model based on ARD kernel function has the highest identification accuracy.
SCIENTIFIC REPORTS
(2022)
Article
Computer Science, Artificial Intelligence
Yi-Qi Hu, Xu-Hui Liu, Shu-Qiao Li, Yang Yu
Summary: AutoML aims to automatically configure learning systems by selecting algorithms and tuning hyperparameters. Previous approaches searched in the joint hyperparameter space, which was inefficient. We propose a cascaded algorithm selection method, using ER-UCB strategy, to achieve the goal of finding the best configuration. We introduce ER-UCB-S and ER-UCB-N algorithms for stationary and non-stationary settings respectively, and empirical studies confirm their effectiveness in synthetic and AutoML tasks.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Article
Computer Science, Artificial Intelligence
Di Wu, Ting Li, Qin Wan
Summary: A hybrid deep kernel incremental extreme learning machine (DKIELM) was proposed in this paper, which utilized improved coyote and beetle swarm optimization methods to enhance efficiency and convergence. Experimental results demonstrated the feasibility and effectiveness of the DKIELM, showing superior performance compared to other ELMs.
COMPLEX & INTELLIGENT SYSTEMS
(2021)
Article
Mathematical & Computational Biology
Xianli Liu, Yongquan Zhou, Weiping Meng, Qifang Luo
Summary: This paper presents a novel regression and classifier called Functional Extreme Learning Machine (FELM), which uses functional equation-solving theory to guide the modeling process. Compared to ELM, FELM has better generalization performance and stability.
MATHEMATICAL BIOSCIENCES AND ENGINEERING
(2023)
Article
Engineering, Civil
Zhong-kai Feng, Wen-jing Niu, Zheng-yang Tang, Yang Xu, Hai-rong Zhang
Summary: A novel evolutionary artificial intelligence model is developed for multiple scales nonstationary hydrological time series prediction, utilizing the cooperation search algorithm to optimize the ELM model's input-hidden weights and biases. Experimental results show that the proposed method outperforms the traditional ELM method in terms of performance evaluation indexes, particularly with significant improvements in both RMSE and MAPE during testing phase, supporting decision-making in water resource system.
JOURNAL OF HYDROLOGY
(2021)
Article
Thermodynamics
Zhikun Gao, Junqi Yu, Anjun Zhao, Qun Hu, Siyuan Yang
Summary: The hybrid prediction model RF-IPWOA-ELM is proposed and demonstrated to have high prediction accuracy and short prediction time for cooling load forecasting in large commercial buildings. It shows effectiveness in predicting cooling loads in different months and with limited training samples, indicating strong generalization ability and importance for energy conservation and management in air conditioning systems.
Article
Engineering, Industrial
Shengyan Li, Hongyan Ma, Yingda Zhang, Shuai Wang, Rong Guo, Wei He, Jiechuan Xu, Zongyuan Xie
Summary: In educational facility interiors, the risk of congestion and trampling during evacuation is a significant safety concern. This article proposes an emergency evacuation risk assessment model based on the improved extreme learning machine (ELM). The model shows an accurate prediction rate of more than 92% and can enable efficient and fast risk assessment.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2023)
Article
Energy & Fuels
Jungin Lee, Olivia J. Cook, Andrea P. Arguelles, Yashar Mehmani
Summary: This study explores the link between infrared spectra and the mechanical response of rocks using machine learning algorithms. By utilizing hyperspectral images, the acoustic velocity, acoustic attenuation coefficient, and X-ray attenuation coefficient of shale specimens were accurately predicted. This work demonstrates the potential to predict the physical properties of rocks using infrared spectra.
Article
Computer Science, Artificial Intelligence
Yong Wang, Kuo-Yi Lin, Shuming Cheng, Li Li
Summary: In this paper, a novel variational quantum extreme learning machine (VQELM) is proposed to address the efficiency issue of processing data with extremely large feature spaces. The VQELM uses a special feature mapping method to achieve nonlinear transformation and outperforms classical ELM in classification and regression tasks in classical and quantum simulations.
Article
Mathematics, Interdisciplinary Applications
Jujie Wang, Quan Cui, Maolin He
Summary: In this study, a novel predicting model is proposed to predict carbon price by combining the advantages of the improved variational mode decomposition algorithm, multiscale entropy algorithm, and the extreme learning machine model improved by the intelligent optimization algorithm. The performance indicators of the proposed model are significantly lower than others, indicating its effectiveness in time series prediction.
CHAOS SOLITONS & FRACTALS
(2022)
Article
Chemistry, Analytical
Nebojsa Bacanin, Catalin Stoean, Miodrag Zivkovic, Dijana Jovanovic, Milos Antonijevic, Djordje Mladenovic
Summary: The extreme learning machine is a fast and efficient model, but its performance heavily depends on the weights and biases within the hidden layer. This study proposes a multi-swarm hybrid optimization approach for determining optimal or near optimal weights and biases, using three swarm intelligence meta-heuristics. The proposed method outperforms other similar approaches in terms of generalization performance.
Article
Engineering, Civil
Jincheng Zhou, Dan Wang, Shahab S. Band, Changhyun Jun, Sayed M. Bateni, M. Moslehpour, Hao-Ting Pai, Chung-Chian Hsu, Rasoul Ameri
Summary: This study aimed to forecast the monthly river discharge time-series of two gauging hydrometric sites on the Missouri River using two machine learning models (XGB and KNN). XGB outperformed KNN in forecasting river flow. Wavelet analysis was incorporated to develop hybrid W-XGB and W-KNN approaches. Two novel hybrid models, XGB-LJA and W-XGB-LJA, were established through the hybridization of XGB with the Levy-Jaya optimization algorithm and simultaneous integration of wavelet analysis and LJA with XGB. The performance of the models was evaluated using RMSE, MAE, MBE, R, and NSE. The best discharge forecasts were obtained using the hybrid WXGB2-LJA and W-XGB4-LJA models.
WATER RESOURCES MANAGEMENT
(2023)
Article
Automation & Control Systems
Keven Rall, David Loker, Chetan P. Nikhare
Summary: Remote manufacturing process monitoring is becoming more popular in industry due to its benefits of increased production, reduced costs, and improved product quality. However, the lack of human interaction with the machine can be a drawback. To address this issue, non-invasive sensors such as microphone arrays can be used to remotely detect manufacturing process parameters. Machine learning algorithms, such as k-nearest neighbor (kNN) and Gaussian support vector machine (Gaussian SVM), can then be used to predict these parameters. The study found that using autocorrelation in the frequency domain yielded the best results for all three machining parameters, with accuracy ranging from 90-100%.
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2023)