Evaluation of Machine Learning approach in flood prediction scenarios and its input parameters: A systematic review

• Main Authors: Maspo, Nur-Adib, Harun, Aizul Nahar, Goto, Masafumi, Cheros, Faizah, Haron, Nuzul Azam, Mohd Nawi, Mohd Nasrun
• Format: Article
• Language: English
• Published: IOP Publishing 2020
• Subjects: HD28 Management. Industrial Management
• Online Access: http://repo.uum.edu.my/28272/1/RCND%202019%201%2014.pdf; http://repo.uum.edu.my/28272/; http://doi.org/10.1088/1755-1315/479/1/012038

Flood disaster is a major disaster that frequently happens globally, it brings serious impacts to lives, property, infrastructure and environment. To stop flooding seems to be difficult but to prevent from serious damages that caused by flood is possible. Thus, implementing flood prediction could help in flood preparation and possibly to reduce the impact of flooding. This study aims to evaluate the existing machine learning (ML) approaches for flood prediction as well as evaluate parameters used for predicting flood, the evaluation is based on the review of previous research articles. In order to achieve the aim, this study is in two-fold; the first part is to identify flood prediction approaches specifically using ML methods and the second part is to identify flood prediction parameters that have been used as input parameters for flood prediction model. The main contribution of this paper is to determine the most recent ML techniques in flood prediction and identify the notable parameters used as model input so that researchers and/or flood managers can refer to the prediction results as the guideline in considering ML method for early flood prediction.