典型文献
Attention-Based Multi-Scale Prediction Network for Time-Series Data
文献摘要：
Time series data is a kind of data accu-mulated over time,which can describe the change of phenomenon.This kind of data reflects the degree of change of a certain thing or phenomenon.The exist-ing technologies such as LSTM and ARIMA are bet-ter than convolutional neural network in time series prediction,but they are not enough to mine the peri-odicity of data.In this article,we perform periodic analysis on two types of time series data,select time metrics with high periodic characteristics,and propose a multi-scale prediction model based on the attention mechanism for the periodic trend of the data.A loss calculation method for traffic time series characteris-tics is proposed as well.Multiple experiments have been conducted on actual data sets.The experiments show that the method proposed in this paper has bet-ter performance than commonly used traffic prediction methods(ARIMA,LSTM,etc.)and 3％-5％increase on MAPE.
文献关键词：
中图分类号：
作者姓名：
Junjie Li;Lin Zhu;Yong Zhang;Da Guo;Xingwen Xia
作者机构：
School of Electronic Engineering,Beijing University of Posts and Telecommunications,Beijing 100876,China;Beijing Key Laboratory of Work Safety Intelligent Monitoring(Beijing University of Posts and Telecommunications),Beijing 100876,China;China Mobile Research Institute,Beijing 100053,China
文献出处：
引用格式：
[1]Junjie Li;Lin Zhu;Yong Zhang;Da Guo;Xingwen Xia-.Attention-Based Multi-Scale Prediction Network for Time-Series Data)[J].中国通信（英文版）,2022(05):286-301
A类：
odicity
B类：
Attention,Based,Scale,Prediction,Network,Time,Series,Data,series,data,kind,accu,mulated,over,which,can,describe,change,phenomenon,This,reflects,degree,certain,thing,exist,technologies,such,ARIMA,are,bet,than,convolutional,neural,network,prediction,but,they,not,enough,mine,In,this,article,periodic,analysis,types,select,metrics,high,characteristics,multi,scale,model,attention,mechanism,trend,loss,calculation,traffic,proposed,well,Multiple,experiments,have,been,conducted,actual,sets,show,that,paper,has,performance,commonly,used,methods,etc,increase,MAPE
AB值：
0.545832
相似文献
机标中图分类号，由域田数据科技根据网络公开资料自动分析生成，仅供学习研究参考。
