Article
Automation & Control Systems
Junhua Zheng, Yingkai Gong, Wei Liu, Le Zhou
Summary: This paper proposes a set of ensemble Gaussian process regression (GPR) models for nonlinear spectroscopic calibration. The new subspace GPR model constructs multiple subspaces along uncorrelated directions to improve the robustness and diversity of the ensemble model. Comparative studies show that the new subspace GPR model improves both prediction accuracy and robustness.
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS
(2022)
Article
Spectroscopy
Hui Chen, Chao Tan, Zan Lin
Summary: This study explored the feasibility of using near-infrared spectroscopy combined with ensemble learning to discriminate the origin of Ginseng. The final models showed high sensitivity, specificity, and overall accuracy.
SPECTROCHIMICA ACTA PART A-MOLECULAR AND BIOMOLECULAR SPECTROSCOPY
(2024)
Article
Green & Sustainable Science & Technology
Rana Muhammad Adnan, Abolfazl Jaafari, Aadhityaa Mohanavelu, Ozgur Kisi, Ahmed Elbeltagi
Summary: The study developed advanced computational models for streamflow forecasting, combining LWL algorithm with ensemble techniques. Incorporating periodicity and cross-validation techniques improved model performance, with dataset M3 providing the most accurate results. The ensemble LWL-AR model outperformed other models, highlighting the robustness of ensemble modeling approach for streamflow forecasting.
Article
Computer Science, Information Systems
Quan Wang, Fei Wang, Zhongheng Li, Peilin Jiang, Fuji Ren, Feiping Nie
Summary: This paper presents a novel framework called Efficient Random Subspace decision forest (ERS), which uses the HRDUVD method to determine the dimensionality of the random subspace. The ERS framework achieves effective and efficient decision forests in high dimensional cases.
INFORMATION SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Yun-Hao Cao, Jianxin Wu, Hanchen Wang, Joan Lasenby
Summary: The paper introduces a novel deep learning based random subspace method, NRS, which outperforms traditional forest methods with better representation learning and higher accuracy, demonstrating superior performance on machine learning datasets and achieving improvements in image and point cloud recognition tasks.
PATTERN RECOGNITION
(2021)
Article
Agriculture, Multidisciplinary
P. Berzaghi, J. H. Cherney, M. D. Casler
Summary: Near infrared reflectance (NIR) spectroscopy has made significant advancements since the 1970s, becoming the most commonly used technique for forage analysis. A study comparing different portable instruments to a laboratory NIR instrument found that alfalfa had better calibration and test-set statistics than grasses. The laboratory instrument performed best, while the SCiO portable instrument had higher error rates, offering potential for on-farm analysis in the future.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2021)
Article
Chemistry, Analytical
Bai Xue, Glenn Cloud, Sergey Vishnyakov, Zubin Mehta, Evan Ramer, Feng Jin, Meiping Song, Chein- Chang
Summary: A novel method for NIR spectroscopy spectra standardization is proposed in this paper. Most existing methods for standardization require spectral data sets from both primary and secondary instruments for validation, which limits their usage. This paper investigates the issue of spectrum data order and develops a different approach based on statistical signal processing. The developed method compensates for distortion and transfers the second order statistic from the primal spectra to the secondary spectra, allowing estimation regardless of the sample statistic order. Application-driven experiments and a comparison to PDS are conducted to demonstrate the extended usage of the method in NIR spectra standardization.
ANALYTICA CHIMICA ACTA
(2023)
Article
Geography, Physical
Binh Thai Pham, Abolfazl Jaafari, Trung Nguyen-Thoi, Tran Van Phong, Huu Duy Nguyen, Neelima Satyam, Md Masroor, Sufia Rehman, Haroon Sajjad, Mehebub Sahana, Hiep Van Le, Indra Prakash
Summary: This study developed highly accurate ensemble machine learning models for spatial prediction of rainfall-induced landslides in the Uttarkashi district, India. The D-REPT model was identified as the most accurate, providing insights for engineers and modelers to develop more advanced predictive models.
INTERNATIONAL JOURNAL OF DIGITAL EARTH
(2021)
Article
Agricultural Engineering
M. I. S. Verissimo, C. Soares, C. F. Moreirinha, M. T. S. R. Gomes
Summary: In this study, a method based on NIR spectroscopy was used to predict the pour point and ethanol content of biodiesel mixtures. The results showed that this method has good prediction capability and is faster and more convenient compared to the traditional ASTM D97-08 procedure.
BIOMASS & BIOENERGY
(2023)
Article
Spectroscopy
Xiao-Wen Zhang, Zheng-Guang Chen, Feng Jiao
Summary: The dimensionality of near-infrared (NIR) spectral data is often large, and dimensionality reduction is crucial for increasing the model's performance. Laplacian Eigenmaps (LE) can preserve local neighborhood information but is disturbed by irrelevant information and multicollinearity. Random Frog (RF) algorithm can eliminate noise and collinearity. Hence, before using LE, RF is used to eliminate irrelevant information and reduce correlation, resulting in improved regression models' prediction accuracy and stability.
SPECTROCHIMICA ACTA PART A-MOLECULAR AND BIOMOLECULAR SPECTROSCOPY
(2023)
Article
Computer Science, Artificial Intelligence
Mohammad Sultan Mahmud, Joshua Zhexue Huang, Salvador Garcia
Summary: This study proposes a new distributed clustering approximation framework for big data, which uses multiple random samples to compute an ensemble result and integrates component clustering results using two new methods. Experimental results demonstrate the accuracy in identifying the correct number of clusters and the better scalability, efficiency, and clustering stability of the proposed methods.
INFORMATION FUSION
(2024)
Article
Engineering, Environmental
Ahmed Elbeltagi, Manish Kumar, N. L. Kushwaha, Chaitanya B. Pande, Pakorn Ditthakit, Dinesh Kumar Vishwakarma, A. Subeesh
Summary: This study examines the feasibility and effectiveness of the Random Subspace (RSS) model and its hybridization with M5 Pruning tree (M5P), Random Forest (RF), and Random Tree (RT) to estimate the Standardized Precipitation Index (SPI) for droughts in Rajasthan, India.
STOCHASTIC ENVIRONMENTAL RESEARCH AND RISK ASSESSMENT
(2023)
Article
Computer Science, Artificial Intelligence
Somayeh Emami, Vahid Rezaverdinejad, Hossein Dehghanisanij, Hojjat Emami, Ahmed Elbeltagi
Summary: Soil water content (SWC) is crucial for water and soil resource management, and accurate prediction of SWC is important in water and soil studies. This research investigates the performance of four data mining algorithms for SWC prediction and finds that the random tree algorithm outperforms others. The developed models in this study can assist agricultural water users, developers, and decision-makers in achieving agricultural sustainability.
Article
Computer Science, Artificial Intelligence
Sampath Deegalla, Keerthi Walgama, Panagiotis Papapetrou, Henrik Bostrom
Summary: The random subspace and random projection methods were investigated for forming ensembles of nearest neighbor classifiers in high dimensional feature spaces, with results showing improvements in predictive performance compared to standard nearest neighbor classifiers. The choice between the two methods depends on the type of data, with random projection outperforming random subspace for microarray and chemoinformatics datasets, while the opposite is true for image datasets. Additionally, the resulting ensembles using random projection perform on par with random forests for microarray and chemoinformatics datasets.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Engineering, Civil
Yahi Takai Eddine, Marouf Nadir, Sehtal Sabah, Abolfazl Jaafari
Summary: This study proposes an ensemble modeling approach that integrates support vector machine with several ensemble learning techniques to predict flow rates in natural rivers of a Mediterranean climate in Algeria. The results indicate that the ensemble models outperform the standalone support vector machine model, with SVM-Dagging model performing the best.
WATER RESOURCES MANAGEMENT
(2023)