A Bayesian Framework Based on Gaussian Mixture Model and Radial Basis Function Fisher Discriminant Analysis for Flood Spatial Prediction (BayGmmKda V1.1)
Dieu Tien Bui1 and Nhat-Duc Hoang21Geographic Information System Group, Department of Business Administration and Computer Science, University College of Southeast Norway (USN), Hallvard Eikas Plass, N-3800, Bø I Telemark, Norway 2Faculty of Civil Engineering, Institute of Research and Development, Duy Tan University, P809 - K7/25 Quang Trung, Danang, Vietnam
Received: 19 Dec 2016 – Accepted for review: 16 Jan 2017 – Discussion started: 17 Jan 2017
Abstract. In this study, a probabilistic model, named as BayGmmKda, is proposed for flood assessment with a study area in Central Vietnam. The new model is essentially a Bayesian framework constructed a combination of Gaussian Mixture Model, Radial Basis Function Fisher Discriminant Analysis, and a Geographic Information System database. Experiments used for measuring the model performance point out that the hybrid framework is superior to other benchmark models including the adaptive neuro fuzzy inference system and the support vector machine. To facility the model implementation, a software program of BayGmmKda has been developed in Matlab environment. The newly proposed model is shown to be a very promising alternative for assisting decision-makers in flood assessment.
Tien Bui, D. and Hoang, N.-D.: A Bayesian Framework Based on Gaussian Mixture Model and Radial Basis Function Fisher Discriminant Analysis for Flood Spatial Prediction (BayGmmKda V1.1), Geosci. Model Dev. Discuss., doi:10.5194/gmd-2016-311, in review, 2017.