基于机器学习的登革热时空扩散预测模型对比分析

Comparing of Spatio-Temporal Diffusion Prediction Models of Dengue Fevers Based on Machine Learning

BP神经网络、GA-BP神经网络及SVR模型是机器学习领域常用的三种预测方法,但在登革热预测方面鲜有人涉及。本文以广州市主城区登革热预测为例,对比BP神经网络、GA-BP神经网络及SVR模型在登革热时空预测上的作用,比较三种模型在登革热时空动态预测中的优劣性。研究表明,1从模型预测效果上看,SVR模型稳定,预测效果显著优于BP及GA-BP模型;2从模型性能上看,GA-BP模型优于BP及SVR模型; 3SVR与GA-BP模型在登革热预测上切实可行。

BP neural network, GA-BP neural network and SVR model are commonly used in the field of machine learning, but few of them are involved in the diffusion prediction of dengue fever. In this paper, we took Dengue Fever in the downtown of Guangzhou city as an example, compared the spatio-temporal dynamics prediction results of BP neuralnetwork, GA-BP neural network and SVR models. The results showed that, the prediction effect of SVR model was superior to BP and GA-BP model; the performance of GA-BP model was better than BP and SVR model; SVR and GABPmodel were feasible in the prediction of Dengue Fever.