Fire Situation Forecasting Based on Support Vector Machine Optimized by Genetic Algorithm

Abstract

Improve the prediction accuracy of fire situation reasonably has great significance for fire prevention and fire deployment. Firstly, build a fire situation prediction model by using support vector regression; followed adopt genetic algorithm to select the optimal combination of parameters; finally provide empirical analysis by taking Chinese Zhejiang Province, test reliability and practicality of model. The results showed that: the fire prediction model based on support vector machine has ideal learning ability and generalization ability; the predicted results possess a high precision, thus providing the new idea and method for predicting fire situation.