Application of Support Vector Machine with Posterior Probability Estimates in Debris Flow Hazard Assessment
Xiuzhen Li, Jiming Kong · 2011
Support Vector Machine (SVM), a new machine learning method based on Statistical Learning Theory, has been widely applied in various fields because of the excellent learning performance and unique advantages in solving small-sample, non-linear and high-dimensional problems. In this paper; we built a multi-class SVM model with posterior probability estimates for hazard assessment of regional debris flow. It is shown by the instances that the SVM model has higher assessment accuracy rate (the accuracy rates for training and testing samples are 90% and 92.86% respectively and the accuracy rate for total samples is 91.18%) and can give a probability belonging to each class as well as a class for debris flow hazard.