首页|University of Sharjah Researchers Describe Recent Advances in Artificial Intelli gence (Drought prediction using artificial intelligence models based on climate data and soil moisture)
University of Sharjah Researchers Describe Recent Advances in Artificial Intelli gence (Drought prediction using artificial intelligence models based on climate data and soil moisture)
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-Research findings on artificial intell igence are discussed in a new report. According to news reporting out of the Uni versity of Sharjah by NewsRx editors, research stated, "Drought is deemed a majo r natural disaster that can lead to severe economic and social implications. Dro ught indices are utilized worldwide for drought management and monitoring." Our news correspondents obtained a quote from the research from University of Sh arjah: "However, as a result of the inherent complexity of drought phenomena and hydroclimatic condition differences, no universal drought index is available fo r effectively monitoring drought across the world. Therefore, this study aimed t o develop a new meteorological drought index to describe and forecast drought ba sed on various artificial intelligence (AI) models: decision tree (DT), generali zed linear model (GLM), support vector machine, artificial neural network, deep learning, and random forest. A comparative assessment was conducted between the developed AI-based indices and nine conventional drought indices based on their correlations with multiple drought indicators. Historical records of five drough t indicators, namely runoff, along with deep, lower, root, and upper soil moistu re, were utilized to evaluate the models' performance. Different combinations of climatic datasets from Alice Springs, Australia, were utilized to develop and t rain the AI models. The results demonstrated that the rainfall anomaly drought i ndex was the best conventional drought index, scoring the highest correlation (0 .718) with the upper soil moisture. The highest correlation between the new and conventional indices was found between the DT-based index and the rainfall anoma ly index at a value of 0.97, whereas the lowest correlation was 0.57 between the GLM and the Palmer drought severity index. The GLM-based index achieved the bes t performance according to its high correlations with conventional drought indic ators, e.g., a correlation coefficient of 0.78 with the upper soil moisture."
University of SharjahArtificial Intell igenceEmerging TechnologiesMachine Learning
NETL
NSTL
