首页|New Machine Learning Study Findings Have Been Published by Researchers at K.N. T oosi University of Technology (Land subsidence susceptibility mapping based on I nSAR and a hybrid machine learning approach)
New Machine Learning Study Findings Have Been Published by Researchers at K.N. T oosi University of Technology (Land subsidence susceptibility mapping based on I nSAR and a hybrid machine learning approach)
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
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – New study results on artificial intell igence have been published. According to news reporting originating from Tehran, Iran, by NewsRx correspondents, research stated, “Land subsidence (LS) due to n atural processes or human activity can irreparably damage the environment. This study employed the quasi-permanent scatterer method to detect areas with and wit hout subsidence in the period from 2018 to 2020.” The news correspondents obtained a quote from the research from K.N. Toosi Unive rsity of Technology: “In addition, 12 factors affecting subsidence were selected to detect LS-prone areas. Information gain ratio (IGR) and frequency ratio meth ods were used to determine the importance and weighting of various factors and s ub-factors affecting subsidence. Novel approaches, including the standard adapti ve-networkbased fuzzy inference system (ANFIS) algorithm and its integration wi th the particle swarm optimization (PSO) algorithm, yielded LS maps. The models’ predictive performance was assessed using statistical indexes such as the root mean square error (RMSE), area under the receiver operating characteristic curve (AUROC) and confusion matrix criteria (e.g., sensitivity, specificity, precisio n, accuracy, and recall). Finally, Shapley additive explanations approach was us ed to explore the importance of features in modeling. The findings showed that t he subsidence pattern was V-shaped, averaging 321 mm/year. Ground-leveling and i nterferometric synthetic aperture radar measurements revealed a 0.93 correlation coefficient with a s = 1.43 mm/year deformation rate. Based on IGR analysis, aq uifer media, the decline of the water table, and aquifer thickness played pivota l roles in LS occurrences. In addition, the ANFIS-PSO model classified approxima tely 50.26 % of the study area as high and very high susceptibilit y. The AUROC values of ANFIS-PSO and ANFIS models for the training dataset were 0.912 and 0.807, respectively. For the testing dataset, the ANFIS-PSO model prod uced a higher AUROC value of 0.863, while the ANFIS model had a value of 0.771.”
K.N. Toosi University of TechnologyTeh ranIranAsiaCyborgsEmerging TechnologiesMachine Learning
NETL
NSTL
