Article
Agriculture, Dairy & Animal Science
G. Rovere, G. de los Campos, A. L. Lock, L. Worden, A. Vazquez, K. Lee, R. J. Tempelman
Summary: Through analyzing a large number of milk samples, the study found that Bayesian regression methods outperformed partial least squares in predicting milk fatty acids, and identified spectral regions associated with fatty acids as well as the impact of carbon number and unsaturation level on the strength of associations.
JOURNAL OF DAIRY SCIENCE
(2021)
Article
Environmental Sciences
Everson Cezar, Tatiane Amancio Alberton, Evandro Freire Lemos, Karym Mayara de Oliveira, Liang Sun, Luis Guilherme Teixeira Crusiol, Marlon Rodrigues, Amanda Silveira Reis, Marcos Rafael Nanni
Summary: The quantification of soil organic matter (SOM) has been increasing in the Brazilian Cerrado region, where SOM content tends to be low. This study evaluated the performance of a local spectral model for SOM prediction using spectroradiometry. The results showed that recalibration of the local models improved the prediction accuracy, but further research is needed to improve the identification of SOM spatial variability.
Article
Agriculture, Dairy & Animal Science
Fang-Li Qin, Xin-Chun Wang, Si-Ran Ding, Guang-Sheng Li, Zhuo-Cheng Hou
Summary: This study successfully developed a fast method for measuring the fat content of Peking duck using near-infrared spectroscopy, which demonstrated good predictive abilities.
Article
Engineering, Geological
Erika Schiappapietra, John Douglas
Summary: This study examines the application of geostatistical methods based on the likelihood function in estimating the spatial correlation of ground motion intensity measures. Maximum-likelihood methods may provide correlation estimates that are not dependent on the bin size and could outperform least-squares methods.
BULLETIN OF EARTHQUAKE ENGINEERING
(2021)
Article
Computer Science, Interdisciplinary Applications
Lavi Rizki Zuhal, Ghifari Adam Faza, Pramudita Satria Palar, Rhea Patricia Liem
Summary: This paper aims to assess the potential of Kriging combined with partial least squares (KPLS) for fast uncertainty quantification and sensitivity analysis in high-dimensional problems. Results show that KPLS with four principal components is significantly faster than the ordinary Kriging while yielding comparable accuracy in approximating the statistical moments and Sobol indices.
STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION
(2023)
Article
Business, Finance
Ziyu Song, Changrui Yu
Summary: A new group of investor sentiment indices constructed using a new dimension reduction technique shows superior ability in predicting market returns and generating economic value for investors.
INTERNATIONAL REVIEW OF FINANCIAL ANALYSIS
(2022)
Article
Optics
Zhenyu Li, Hui Zhang, Binh Thi Thanh Nguyen, Shaobo Luo, Patricia Yang Liu, Jun Zou, Yuzhi Shi, Hong Cai, Zhenchuan Yang, Yufeng Jin, Yilong Hao, Yi Zhang, Ai-Qun Liu
Summary: The smart sensor achieves label-free multicomponent chemical analysis with a high prediction accuracy and simple detection strategy using a label-free ring resonator and neural network model, promising great potential for applications in various fields.
PHOTONICS RESEARCH
(2021)
Article
Spectroscopy
Pauline Ong, Suming Chen, Chao-Yin Tsai, Yung-Kun Chuang
Summary: In this study, near-infrared spectroscopy was used for non-destructive determination of theanine in oolong tea. The flower pollination algorithm (FPA) was utilized for selecting discriminative wavelengths for partial least squares regression (PLSR), resulting in improved predictive performance and reduced model complexity. The optimized model using FPA outperformed other wavelength selection methods and showed potential for enhancing predictability of PLSR.
SPECTROCHIMICA ACTA PART A-MOLECULAR AND BIOMOLECULAR SPECTROSCOPY
(2021)
Article
Engineering, Mechanical
Yushan Liu, Luyi Li, Sihan Zhao
Summary: This paper combines Kriging and PLS to develop a global surrogate model for high-dimensional structural systems, named as PLS-K. In the proposed method, PLS is employed to identify the input-output principal components, wherein Kriging model is used to establish the relationship between each pair of principal components.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Multidisciplinary Sciences
Xiangyu Guo, Ahmed Jahoor, Just Jensen, Pernille Sarup
Summary: In this study, metabolomic spectra were used to predict malting quality phenotypes in different locations, and the prediction ability of different models and training population sizes were compared. The results showed that more than 90% of the total variance in malting quality traits could be explained by metabolomic features. The prediction accuracy increased with increasing training population size and stabilized when the size reached 1000. The optimal number of components considered in the prediction models was 20. The accuracy using cross-validation ranged from 0.722 to 0.865 for leave-one-line-out and from 0.517 to 0.817 for leave-one-location-out. Therefore, metabolomic prediction of malting quality traits using metabolomic features has high accuracy, and MBLUP is better than PLSR when the training population size is larger than 100.
SCIENTIFIC REPORTS
(2022)
Article
Engineering, Manufacturing
Stephano Piva, Andre Nogueira Assis, Petrus Christiaan Pistorius, Michael Kan
Summary: A method combining statistics and process engineering using partial least squares regression (PLS) is developed to predict non-metallic inclusion content and composition in steel. The model can accurately predict total oxygen content and Mg/(Mg+Al) ratio, providing enough data to recommend Ca addition based on thermodynamic calculation. It can also predict total inclusion fraction and accurately predict average CaS content and Ca/Al ratio in the tundish. However, model interpretability is hindered by high dimensionality and multicollinearity of the data. Non-metallic inclusion compositions correspond to the expected composition at the onset of CaS formation based on steel composition.
INTEGRATING MATERIALS AND MANUFACTURING INNOVATION
(2023)
Article
Mathematics
Hongmei Shi, Xingbo Zhang, Yuzhen Gao, Shuai Wang, Yufu Ning
Summary: This article introduces an uncertain total least squares estimation method and an uncertain robust total least squares linear regression method based on uncertainty theory and total least squares method. These methods can handle non-random and imprecise data and achieve more reasonable fitting effect and reliable results.
Article
Geosciences, Multidisciplinary
Masoud Davari, Salah Aldin Karimi, Hossein Ali Bahrami, Sayed Mohammad Taher Hossaini, Soheyla Fahmideh
Summary: The study demonstrated that using spectroscopy for rapid estimation of soil engineering properties showed promising results in different soil samples, with good accuracy for some properties but room for improvement for others.
Article
Engineering, Chemical
Gergo Ignacz, Gyorgy Szekely
Summary: Methods for determining solute rejection in organic solvent nanofiltration are time-consuming and expensive. This study presents two prediction methods based on quantitative structure-activity relationship using traditional machine learning and deep learning models, providing a new platform for a more in-depth investigation into membrane-solute interactions.
JOURNAL OF MEMBRANE SCIENCE
(2022)
Article
Automation & Control Systems
Louna Alsouki, Laurent Duval, Clement Marteau, Rami El Haddad, Francois Wahl
Summary: Relating variables X to response y is important in chemometrics. Qualitative interpretation can enhance quantitative prediction by identifying influential features. Projections (e.g. PLS) and variable selections (e.g. lasso) are used for dimension reduction in high-dimensional problems. Dual-sPLS, a variant of PLS1, provides a balance between accurate prediction and efficient interpretation through penalizations inspired by classical regression methods and the dual norm notion. It performs favorably compared to similar regression methods on simulated and real chemical data.
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS
(2023)
Article
Soil Science
Cynthia C. E. van Leeuwen, Vera L. Mulder, Niels H. Batjes, Gerard B. M. Heuvelink
Summary: There is a growing demand for high-quality soil data, with this study quantifying uncertainties in wet chemistry soil data using a linear mixed-effects model. Experimental measurement design and replicate measurements were found to be crucial for accurate uncertainty quantification in soil data.
EUROPEAN JOURNAL OF SOIL SCIENCE
(2022)
Article
Soil Science
Luc Steinbuch, Dick J. Brus, Gerard B. M. Heuvelink
Summary: This study aimed to evaluate if extending a Bayesian Generalized Linear Model (BGLM) to a Bayesian Generalized Linear Geostatistical Model (BGLGM) is worth it for mapping binary soil properties. The results showed that BGLGM performs considerably better than BGLM in terms of statistical validation metrics when a large observation sample and few relevant covariates are available, although it is more demanding in terms of calibration and application.
EUROPEAN JOURNAL OF SOIL SCIENCE
(2022)
Article
Agronomy
Rafael A. Segura, Jetse J. Stoorvogel, Jorge A. Sandoval
Summary: The response of soil management, particularly in terms of pH and nitrogen fertilization, on Fusarium wilt in Gros Michel banana varied significantly among different soil types. This complicates the development of standard soil management strategies for mitigating or combating the disease.
Article
Multidisciplinary Sciences
Gifty E. Acquah, Javier Hernandez-Allica, Cathy L. Thomas, Sarah J. Dunham, Erick K. Towett, Lee B. Drake, Keith D. Shepherd, Steve P. McGrath, Stephan M. Haefele
Summary: The increasing popularity of local blending of fertilisers raises concerns about quality control, fertiliser adulteration, and contamination with trace elements. This study developed empirical calibrations for a portable handheld X-ray fluorescence (pXRF) spectrometer and demonstrated its accuracy in measuring macro and micro nutrients, as well as trace elements and potential contaminants in fertilisers. The results show that pXRF can provide reliable and efficient detection of these substances.
Article
Geosciences, Multidisciplinary
Gerard B. M. Heuvelink, Richard Webster
Summary: Pedologists traditionally mapped soil by drawing boundaries, but the introduction of geostatistics and ordinary kriging in the 1980s revolutionized soil mapping. Machine learning techniques have also been adopted, but they lack transparency and spatial correlation considerations. Spatial statisticians and pedometricians have important roles in incorporating uncertainty and communicating it to end users.
SPATIAL STATISTICS
(2022)
Article
Environmental Sciences
Maria Eliza Turek, Laura Poggio, Niels H. Batjes, Robson Andre Armindo, Quirijn de Jong van Lier, Luis de Sousa, Gerard B. M. Heuvelink
Summary: The development of point-based global maps of soil water retention improves the availability and quality of soil data, compared to traditional map-based approaches. By combining measured and predicted data with environmental variables, this study demonstrates the superior performance of the point-based mapping approach.
INTERNATIONAL SOIL AND WATER CONSERVATION RESEARCH
(2023)
Article
Soil Science
M. E. Angelini, G. B. M. Heuvelink, P. Lagacherie
Summary: This study aimed to help urban planners preserve soils of the highest quality by mapping a soil potential multifunctionality index. However, the prediction accuracy was poor and further improvements are needed.
EUROPEAN JOURNAL OF SOIL SCIENCE
(2023)
Article
Soil Science
Stephan van der Westhuizen, Gerard B. M. Heuvelink, David P. Hofmeyr
Summary: In digital soil mapping, traditional univariate methods neglect the dependence structure between soil properties, while multivariate machine learning models can capture complex non-linear relationships and maintain the dependence structure. This study compares the performance of a multivariate random forest model with two separate univariate random forest models, and finds that the multivariate model outperforms in maintaining the dependence structure and producing more realistic results.
Article
Environmental Sciences
Powell Mponela, Ermias Aynekulu, Mohammed Ebrahim, Tsion Abate, Wuletawu Abera, Haley Zaremba, Marlene Elias, Lulseged Tamene
Summary: The importance of land restoration has gained global attention, but gender-blind policies and strategies exacerbate social inequalities. This study aimed to capture the differences in men and women's perceptions of ecosystem services in Ethiopia before and after restoration interventions. Findings showed a significant disagreement between men's and women's ratings of restoration outcomes, with men attributing degradation to landscape conditions and natural forces, while women identified the lack of appropriate restoration strategies as a precursor for accelerated degradation. The study also revealed that men tended to benefit more from enhanced ecosystem services post-restoration, leading to increased burdens on women in terms of labor and land management.
LAND DEGRADATION & DEVELOPMENT
(2023)
Article
Soil Science
Eyasu Elias, Peter F. Okoth, Jetse J. Stoorvogel, Gezahegn Berecha, Beyene Teklu Mellisse, Abate Mekuriaw, Girmay Gebresamuel, Yihenew G. Selassie, Gizachew Kebede Biratu, Eric M. A. Smaling
Summary: To ensure national food security, there is an urgent need to increase cereal yields in the Ethiopian Highlands. A crop response-to-fertilizer program was conducted as part of the CASCAPE project from 2017 to 2019, focusing on maize, teff, and wheat in different soil groups. Multiple nutrient fertilizers were applied, resulting in varying yields and correlations with soil types and organic carbon levels. The use of "recommendation windows" is suggested for developing fertilizer use policies at district levels, combined with soil maps.
NUTRIENT CYCLING IN AGROECOSYSTEMS
(2023)
Article
Ecology
Alexandre M. J. -C. Wadoux, Gerard B. M. Heuvelink
Summary: Global, continental and regional maps of natural resources are important for assessing ecosystem response to human disturbance and global warming. However, these maps suffer from multiple error sources, and it is important to report the associated uncertainties for users to evaluate their reliability.
METHODS IN ECOLOGY AND EVOLUTION
(2023)
Review
Environmental Sciences
Pieter Jan Kole, Frank G. A. J. Van Belleghem, Jetse J. Stoorvogel, Ad M. J. Ragas, Ansje J. Lohr
Summary: Tyre granulate used as infill for artificial turf is controversial, with some considering it a good example of reuse and others seeing it as a harmful way to dispose of discarded tyres. Due to loose application, the particles disperse into the environment, raising concerns about environmental and health impacts. This study aims to identify the pathways through which infill leaves a football turf and estimate the quantity of infill leaving the turf.
SCIENCE OF THE TOTAL ENVIRONMENT
(2023)
Article
Agriculture, Multidisciplinary
Morgane Batkai, Jean Huge, Dave Huitema, Janjaap Semeijn, Wim Lambrechts, Jetse Stoorvogel
Summary: Building resilient agricultural systems requires transformative action, and social learning is emerging as a promising mechanism for inspiring these transformations. However, there is limited understanding of how or why learning leads to the adoption of transformative practices. While social learning has a positive impact on participants' understanding and adoption of resilient agricultural practices, sustained adoption of transformative actions remains a challenge.
NJAS-IMPACT IN AGRICULTURAL AND LIFE SCIENCES
(2023)
Article
Water Resources
Nega Chalie Emiru, John Walker Recha, Julian R. Thompson, Abrham Belay, Ermias Aynekulu, Alen Manyevere, Teferi D. Demissie, Philip M. Osano, Jabir Hussein, Mikias Biazen Molla, Girma Moges Mengistu, Dawit Solomon
Summary: This study investigated the impacts of climate change on the hydrology of the Upper Awash Basin in Ethiopia. The study found that future temperature will increase, and there will be changes in precipitation and streamflow, leading to an increase in drought frequency and severity.