VAECGAN: a generating framework for long-term prediction in multivariate time series

Cybersecurity

article

Yin, Xiang¹  Han, Yanni¹  Xu, Zhen¹  Liu, Jie² 

[1] Institute of Information Engineering, Chinese Academy of Sciences, School of Cyber Security, University of Chinese Academy of Sciences;Network information department, China Mobile Communications Group Co., Ltd

关键词: Long-term prediction;    Multivariate time series;    Attention mechanism;    Generating framework;   

DOI  :  10.1186/s42400-021-00090-w

学科分类：社会科学、人文和艺术（综合）

来源: Springer

【 摘 要 】

Long-term prediction is still a difficult problem in data mining. People usually use various kinds of methods of Recurrent Neural Network to predict. However, with the increase of the prediction step, the accuracy of prediction decreases rapidly. In order to improve the accuracy of long-term prediction,we propose a framework Variational Auto-Encoder Conditional Generative Adversarial Network(VAECGAN). Our model is divided into three parts. The first part is the encoder net, which can encode the exogenous sequence into latent space vectors and fully save the information carried by the exogenous sequence. The second part is the generator net which is responsible for generating prediction data. In the third part, the discriminator net is used to classify and feedback, adjust data generation and improve prediction accuracy. Finally, extensive empirical studies tested with five real-world datasets (NASDAQ, SML, Energy, EEG,KDDCUP)demonstrate the effectiveness and robustness of our proposed approach.

【 授权许可】

CC BY