首页|Report Summarizes Machine Learning Study Findings from Wichita State University (Flood susceptibility mapping: Integrating machine learning and GIS for enhanced risk assessment)
Report Summarizes Machine Learning Study Findings from Wichita State University (Flood susceptibility mapping: Integrating machine learning and GIS for enhanced risk assessment)
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-A new study on artificial intelligence is now ava ilable. According to news reporting from Wichita State University by NewsRx jour nalists, research stated, "Flooding presents a formidable challenge in the Unite d States, endangering lives and causing substantial economic damage, averaging a round $5 billion annually. Addressing this issue and improving comm unity resilience is imperative." Financial supporters for this research include Wichita State University. Our news editors obtained a quote from the research from Wichita State Universit y: "This project employed machine learning techniques and publicly available dat ato explore the factors influencing flooding and to develop flood susceptibilit y maps at various spatial resolutions. Six machine learning algorithms, includin g Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), K- nearest neighbor (KNN), Adaptive Boosting (Ada Boost), and Extreme Gradient Boos ting (XGB) were used. Geospatial datasets comprising thirteen predictor variable s and 1528 flood inventory data collected since 1996 were analyzed. The predicto r variables are rainfall, elevation, slope, aspect, flow direction, flow accumul ation, Topographic Wetness Index (TWI), distance from the nearest stream, evapot ranspiration, land cover, impervious surface, land surface temperature, and hydr ologic soil group. Five hundred twentyeight non-flood data points were randomly created using a stream buffer for two scenarios. atotal of 2964 data points we re classified into flooded (1) and non-flooded (0) categories and used as atarg et. Overall, testing results showed that the XGB and RF models performed relativ ely well in both cases over multiple resolutions compared to other models, with an accuracy ranging from 0.82 to 0.97. Variable importance analysis depicted that predictor variables such as distance from the streams, hydrologic soil type, r ainfall, elevation, and impervious surfaces significantly affected flood predict ion, suggesting a strong association with the underlying driving process."
Wichita State UniversityAlgorithmsCy borgsEmerging TechnologiesMachine Learning
NETL
NSTL
