Publication: Time Series Water Level Forecasting Based On Convolutional Neural Network (CNN)
Loading...
Date
2020-02
Authors
Nurul Syafiqah binti Zaini
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
In Malaysia, floods are a regular natural disaster that occurs during the certain months of the year, usually from October to March due to the monsoon season. Floods occur when the water level rises above the danger stages due to heavy rainfall. Time series forecasting has been performing many years ago and various methods have been study to test their ability. In this study, Convolutional Neural Network (CNN) is suggested to forecast the time series water level. Two types of scenarios are proposed to study the reliability of the proposed model for water level forecasting. Another two methods are proposed namely Artificial Neural Network (ANN) and Long – Short Term Memory (LSTM) for comparison. In this study, result has confirm that LSTM model can achieve better prediction than the proposed model, in terms of Coefficient of determination (R²) and Mean Square Error (MSE) performance measures
Description
FYP Sem 2 2019/2020
Keywords
Time Series Forecasting , Convolutional Neural Network (CNN)