首页|New Findings on Machine Learning Described by Investigators at Federal Universit y Para (S-wave Log Construction Through Semisupervised Regression Clustering Us ing Machine Learning: a Case Study of North Sea Fields)
New Findings on Machine Learning Described by Investigators at Federal Universit y Para (S-wave Log Construction Through Semisupervised Regression Clustering Us ing Machine Learning: a Case Study of North Sea Fields)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
Researchers detail new data in Machine Learning. According to news reporting originating from Belem, Brazil, by NewsRx correspondents, research stated, "Accurate prediction of S-wave velocity from w ell logs is essential for understanding subsurface geological formations and hyd rocarbon reservoirs. Machine learning techniques, including clustering and regre ssion, have emerged as effective methods for indirectly estimating S-wave logs a nd other rock properties." Financial support for this research came from University of Para<acute accent>. Our news editors obtained a quote from the research from Federal University Para , "In this study, we employed clustering algorithms to identify similarities amo ng well log datasets, encompassing depth, sonic, porosity, neutron, and apparent density, facilitating the discovery of correlations among various wells. These identified correlations served as a foundation for predicting S-wave values usin g a novel semisupervised approach. Our approach combined clustering, specifical ly k-means clustering, with different types of regressors, including Least Squar es Regression (LSR), Support Vector Regression (SVR), and Multi-Layer Perceptron (MLP). Our results demonstrate the superior performance of this integrated appr oach compared to traditional regression methods. We validated our methodology us ing various parametric and non-parametric regression techniques, showcasing its effectiveness not only on wells within the training region but also on wells out side the study area. We achieved a significant improvement in the R2 score metri c, ranging from 2.22% to 6.51%, and a reduction in Me an Square Error (MSE) of at least 31% when compared to predictions made without clustering."
BelemBrazilSouth AmericaCyborgsE merging TechnologiesMachine LearningFederal University Para
NETL
NSTL
