Dynamic risk identification safety model based on fuzzy support vector machine and immune optimization algorithm

Abstract

Dynamic risk identification is used to predict future risk based on known risk information as early as possible. Risk identification technology based on the fuzzy support vector machine (FSVM) model can more comprehensively and automatically identify possible risks through a learning model, and this technology has become the main method for dynamic risk identification. To improve the efficiency and accuracy of recognition, the selection of parameters of the FSVM is very important. An artificial immune algorithm (AIA), an effective stochastic global optimization technology, has advantages of high accuracy and fast convergence speed and does not easily fall into a local optimal solution. Therefore, this paper proposes a new dynamic risk identification model based on the integration of the FSVM and immune optimization algorithm (IOA). By combining the advantages of multi-class FSVM classification and using an artificial IOA to select the FSVM parameters, experiments on the Heart-Disease data set are conducted to demonstrate the effectiveness of the model.