Vol.:(0123456789)
Natural Hazards
https://doi.org/10.1007/s11069-021-04821-7
1 3
ORIGINAL PAPER
Flood‑prone area mapping using machine learning
techniques: a case study of Quang Binh province, Vietnam
Chinh Luu
1
· Quynh Duy Bui
2
· Romulus Costache
3
· Luan Thanh Nguyen
4
·
Thu Thuy Nguyen
5
· Tran Van Phong
6
· Hiep Van Le
7
· Binh Thai Pham
7
Received: 9 November 2020 / Accepted: 24 May 2021
© The Author(s), under exclusive licence to Springer Nature B.V. 2021
Abstract
Vietnam’s central coastal region is the most vulnerable and always at food risk, severely
afecting people’s livelihoods and socio-economic development. In particular, Quang Binh
province is often afected by foods and storms over the year. However, it still lacks studies
on food hazard estimation and prediction tools in this area. This study aims to develop a
fooding susceptibility assessment tool using various machine learning (ML) techniques
namely alternating decision tree (AD Tree), logistic model tree (LM Tree), reduced-error
pruning tree (REP Tree), J48 decision tree (J48) and Naïve Bayes tree (NB Tree); his-
torical food marks; and available data of topography, hydrology, geology, and environ-
ment considering Quang Binh province as a study area. We used food mark locations
of major fooding events in the years 2007, 2010, and 2016; and ten food conditioning
factors to construct and validate the ML models. Various validation methods, including
area under the ROC curve (AUC), were used to validate and compare the models. The
result of the models’ validation suggests that all models have good performance: AD Tree
(AUC = 0.968), LM Tree (AUC = 0.967), REP Tree (AUC = 0.897), J48 (AUC = 0.953),
and NB Tree (AUC = 0.986). Out of these, NB Tree managed to achieve the best perfor-
mance in terms of food prediction with an accuracy higher than 92 %. The fnal food sus-
ceptibility map highlights 6,265 km
2
(78.8 % area) with a very low fooding hazard, 391
km
2
(4.9 % area) with a low fooding hazard, 224 km
2
(2.8 % area) with a moderate fooding
hazard, 243 km
2
(3.1 %) with a high fooding hazard, and 829 km
2
(10.4 % area) with very
high fooding hazard. The fnal fooding susceptibility assessment map could add a valu-
able source for food risk reduction and management activities of Quang Binh province.
Keywords Flood susceptibility map · Alternating decision tree · Logistic model tree ·
Reduced-error pruning tree · J48 · Naïve Bayes tree
* Chinh Luu
luuthidieuchinh@nuce.edu.vn
* Quynh Duy Bui
quynhbd@nuce.edu.vn
* Binh Thai Pham
binhpt@utt.edu.vn
Extended author information available on the last page of the article