NORMA eResearch @NCI Library

Classifying Flood Severity Using Machine Learning

Behera, Jayanta (2020) Classifying Flood Severity Using Machine Learning. Masters thesis, Dublin, National College of Ireland.

[img]
Preview
PDF (Master of Science)
Download (1MB) | Preview
[img]
Preview
PDF (Configuration manual)
Download (860kB) | Preview

Abstract

Flood is one of the most devastating natural hazards that cause huge loss to human life and property. An early and accurate disaster prediction is helpful to prevent the damage. The complexity of factors contributing to flood prediction becomes a challenge in predicting its severity. This research illustrates a novel technique of combining the historical flood incidents with the meteorological and topographic features to predict flood severity by classifying its risk as high, low or moderate. To achieve this, random forest classifier is implemented along with support vector machine, k nearest neighbour, ensemble techniques and neural network. Each of the model is optimized and evaluated based on accuracy, pression, recall and F1-score where random forest classifier outperformed all other techniques with 83% accuracy. This novel technique of combination of historic data with climatic and topographic details showed potential improvement in predicting such catastrophic event which would help in planning proper evacuation and preventing loss of life and property.
Keywords: Flood severity, Random forest, Bagging, accuracy, precision, recall

Item Type: Thesis (Masters)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science

Q Science > QA Mathematics > Computer software
T Technology > T Technology (General) > Information Technology > Computer software
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Dan English
Date Deposited: 18 Jan 2021 16:22
Last Modified: 18 Jan 2021 16:22
URI: http://norma.ncirl.ie/id/eprint/4383

Actions (login required)

View Item View Item