多模型集成的突发传染病预测与可视化平台

Multi-model Integrated Sudden Infectious Disease Prediction and Visualization Platform

为了给各级政府及时应对突发传染病提供一个指导的方案,本文使用了一种改良易感-感染型+长短期记忆网络+自然语言处理的模型对疫情进行预测。模型应用于猴痘感染数据集时,取得了很好的效果,平均绝对百分比误差为21.23%,平均绝对误差为88.03。此外,借助词频-逆文件频率算法和双向长短时神经记忆网络算法,实现了热词云、情感分析等功能。实验结果显示,本文提出的系统能够更准确地分析病毒的传播规律和发展趋势,对相关舆论、新闻等民间信息更敏感,为预测未来突发传染病的传播规律和发展趋势提供了有效的方法。

In order to provide a guiding plan for governments at all levels to respond to sudden infectious diseases in a timely manner,this paper uses an improved susceptible-infected+long short-term memory network+natural language processing model to predict the epidemic situation.When the model was applied to the data set of monkeypox infection,good results were achieved,with an average absolute percentage error of 21.23%and a mean absolute error of 88.03.In addition,with the help of word frequency inverse file frequency algorithm and bidirectional long short term neural memory network algorithm,functions such as hot word cloud and sentiment analysis have been achieved.The experimental results show that the system proposed in this article can more accurately analyze the transmission patterns and development trends of viruses,and is more sensitive to relevant public opinion,news,and other folk information.It provides an effective method for predicting the transmission patterns and development trends of future sudden infectious diseases.