基于LSTM方法的新冠肺炎确诊人数预测模型

LSTM-Based Prediction Model for COVID-19 Confirmed Cases

新冠肺炎疫情的爆发对全球卫生安全、经济发展等方面产生了巨大的影响。因此,构建一个准确的疫情预测模型对于疫情的防控是至关重要的。本次研究基于LSTM时间序列预测模型对美国新冠肺炎疫情期间每日累计确诊病例进行建模和预测。模型包括一个LSTM层,用来捕捉时间序列数据中的长期依赖关系;一个全连接层,用来将模型结果降为一维数值并输出。通过Adam优化器优化模型参数,并且采用均方根对数误差作为模型的评价指标。结果表明,LSTM时间序列预测模型很好地捕捉到了时间序列数据中的长期依赖关系,具备良好的非线性表征能力,较为准确地对美国新冠肺炎疫情发展情况进行预测。

The outbreak of the COVID-19 pandemic has an exerted significant impacts on global health security,economic development,and various other aspects.Consequently,constructing an accurate pandemic prediction model is crucial for effective epidemic prevention and control.An LSTM(Long Short-Term Memory)time series forecasting model was proposed to model and predict the daily cumulative confirmed cases during the COVID-19 outbreak in the United States.And an LSTM layer was incorporated by the model to capture long-term dependencies in time series data and a fully connected layer to reduce the model's output to a one-dimensional value.The model para-meters were optimized using the Adam optimizer,and the root mean square logarithmic error was employed as the evaluation metric.The results demonstrate that the LSTM time series prediction model effectively captures the long-term dependencies in time series data,possesses excellent nonlinear representation capabilities,and accurately predicts the development of the COVID-19 pandemic in the United States.