Flood Risk Prediction in Southern Louisiana Using Support Vector Machines

Authors

Claire Dorsett, Louisiana Tech UniversityFollow

Document Type

Article

Publication Date

Spring 2022

Abstract

Flooding in Southern Louisiana is a growing concern as violent weather becomes more frequent. According to the Environmental Protection Agency (EPA), Louisiana soils have become drier and annual rainfall trends have increased, which may lead to more severe flooding in the coming years. In light of this change in weather trends, flood prediction has become increasingly important for the public’s safety. The focus of this paper is to apply the Support Vector Machine (SVM), a machine learning technique, to classify flood risks based on water gage height, wind speed and direction, and time of the year. In this paper, the methodology of this technique is discussed, and the Support Vector Machine method is applied on the data collected by the National Weather Service (NWS) and the US Army Corps of Engineers using the R language. Finally, this paper will examine the results to discern if this technique is a valid choice for flood risk prediction.

Recommended Citation

Dorsett, Claire, "Flood Risk Prediction in Southern Louisiana Using Support Vector Machines" (2022). Mathematics Senior Capstone Papers. 30.
https://digitalcommons.latech.edu/mathematics-senior-capstone-papers/30