SLIIT >
NCTM - SLIIT >
NCTM - SLIIT 2016 >

Please use this identifier to cite or link to this item: http://hdl.handle.net/123456789/382

Title: Disaster Forecaster
Authors: Karunarathne, D.
Ranawaka, S.
Karunarathne, G.
Thilakakarathna, M.
Tissera, M.
Keywords: Data Mining Algorithms
prediction
classification
Neural Network
Support Vector Machine
Linear Regression
Classification and Regression tree
Issue Date: 2016
Publisher: SLIIT
Abstract: Disaster forecasting is the most challenging problem in all over the world because it consists of multidimensional and nonlinear data. Mostly, weather change causes much trouble in disaster forecasting. This research is based on four data mining algorithms namely Neural Network (NN), Support Vector Machine (SVM), Linear Regression and Decision Tree (DT). Generally these algorithms are used for the forecasting purposes. In this research, historical data on weather for past 10 years has been used covering all the districts in Sri Lanka. The major natural disasters recorded in the history such as floods, droughts, landslides and cyclones are obviously influenced by the extreme weather conditions. These weather conditions which can be measured by the parameters such as rainfall, temperature, humidity and wind are predicted and based on such predictions, main disasters are forecasted with 60% accuracy level in this research.
URI: http://hdl.handle.net/123456789/382
Appears in Collections:NCTM - SLIIT 2016

Files in This Item:

File Description SizeFormat
Disaster Forecaster.pdf263.93 kBAdobe PDFView/Open
View Statistics

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

 

Valid XHTML 1.0! DSpace Software Copyright © 2002-2010  Duraspace - Feedback