Article title: A Comparison of Support Vector Machines and Bayesian Algorithms for Landslide Susceptibility Modeling
Journal information: Geocarto International (United Kingdom) - ISI journal (IF, 2016 = 1.646)
First author: Binh Thai Pham, Co-Head of Geotechnical Engineering and Artificial Intelligence research group, University of Transport Technology, Viet Nam.
Corresponding author: Dieu Tien Bui, Head of Geotechnical Engineering and Artificial Intelligence research group, University of Transport Technology, Viet Nam; Assoc Prof, University College of Southeast Norway, Norway.
Co-authors: Indra Prakash; Khabat Khosravi; Kamran Chapi; Phan Trong Trinh; Trinh Quoc Ngo (GEOAI); Seyed Vrya Hosseini.
In this study, the main goal is to compare the predictive capability of Support Vector Machines (SVM) with four Bayesian algorithms namely Naïve Bayes Tree (NBT), Bayes network (BN), Naïve Bayes (NB), Decision Table Naïve Bayes (DTNB) for identifying landslide susceptibility zones in Pauri Garhwal district (India). Firstly, landslide inventory map was built using 1295 historical landslide data, then in total sixteen influencing factors were selected and tested for landslide susceptibility modeling. Performance of the model was evaluated and compared using Satistical based index methods, Area Under the Receiver Operating Characteristic (ROC) curve named AUC, and Chi-square method. Analysis results show that that the SVM has the highest prediction capability, followed by the NBT, DTNBT, BN and NB, respectively. Thus, this study confirms that the SVM is one of the benchmark models for the assessment of susceptibility of landslides.
The link to the original source: click here