Time Series Prediction Based on Graph Neural Networks
WU Hao
LONG Yin
ZHANG Jinbo
Abstract：In recent years,people have paid more and more attention to artificial intelligence,machine learning algorithms for prediction of time series data have become more and more popular.However,due to the instability of time series data and the cor-relation between series,the results of traditional machine learning methods for time series data are not so good.Therefore,this pa-per proposes a graph neural network prediction model,which takes time series data as input to predict the future closing price,tem-perature,and air quality respectively,and then calculates the prediction based on each prediction.The statistical indicators under the model,MAE,RMSE,are compared with the prediction results of other models.The results show that the graph neural network(GCN-LSTM)prediction model has smaller prediction errors than other(TPA-LSTM,LSTM,Arima,SVR)prediction models,the results are more accurate.
Keywords：GCNtarget valueforecast
Publication Date：2025-06-20
Online Publishing Date：2025-09-23(First online date of this platform, not the publication date of the document)
Pages：5( 1581-1584,1597 )