A Review of Deep Learning Models for Time Series Prediction

被引:280
作者
Han, Zhongyang [1 ]
Zhao, Jun [1 ]
Leung, Henry [2 ]
Ma, King Fai [2 ]
Wang, Wei [1 ]
机构
[1] Dalian Univ Technol, Sch Control Sci & Engn, Dalian 116023, Peoples R China
[2] Univ Calgary, Dept Elect & Comp Engn, Calgary, AB T2N 1N4, Canada
关键词
Time series analysis; Predictive models; Deep learning; Biological system modeling; Support vector machines; Artificial neural networks; Data models; Review; discriminative models; generative models; deep learning; time series prediction; SUPPORT VECTOR MACHINES; EMPIRICAL WAVELET TRANSFORM; RECURRENT NEURAL-NETWORKS; STEP-AHEAD PREDICTION; LONG-TERM PREDICTION; ECHO STATE NETWORK; GAUSSIAN PROCESS; BELIEF NETWORK; SPATIOTEMPORAL PREDICTION; ENSEMBLE APPROACH;
D O I
10.1109/JSEN.2019.2923982
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
In order to approximate the underlying process of temporal data, time series prediction has been a hot research topic for decades. Developing predictive models plays an important role in interpreting complex real-world elements. With the sharp increase in the quantity and dimensionality of data, new challenges, such as extracting deep features and recognizing deep latent patterns, have emerged, demanding novel approaches and effective solutions. Deep learning, composed of multiple processing layers to learn with multiple levels of abstraction, is, now, commonly deployed for overcoming the newly arisen difficulties. This paper reviews the state-of-the-art developments in deep learning for time series prediction. Based on modeling for the perspective of conditional or joint probability, we categorize them into discriminative, generative, and hybrids models. Experiments are implemented on both benchmarks and real-world data to elaborate the performance of the representative deep learning-based prediction methods. Finally, we conclude with comments on possible future perspectives and ongoing challenges with time series prediction.
引用
收藏
页码:7833 / 7848
页数:16
相关论文
共 228 条
[1]   Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI) [J].
Adadi, Amina ;
Berrada, Mohammed .
IEEE ACCESS, 2018, 6 :52138-52160
[2]   Comparison of ARIMA and Artificial Neural Networks Models for Stock Price Prediction [J].
Adebiyi, Ayodele Ariyo ;
Adewumi, Aderemi Oluyinka ;
Ayo, Charles Korede .
JOURNAL OF APPLIED MATHEMATICS, 2014,
[3]  
Al-Qahtani FS, 2013, 2013 INTERNATIONAL CONFERENCE ON COMPUTING, MANAGEMENT AND TELECOMMUNICATIONS (COMMANTEL), P1, DOI 10.1109/ComManTel.2013.6482355
[4]   A new time invariant fuzzy time series forecasting method based on particle swarm optimization [J].
Aladag, Cagdas Hakan ;
Yolcu, Ufuk ;
Egrioglu, Erol ;
Dalar, Ali Z. .
APPLIED SOFT COMPUTING, 2012, 12 (10) :3291-3299
[5]   Short-term prediction of wind power using EMD and chaotic theory [J].
An, Xueli ;
Jiang, Dongxiang ;
Zhao, Minghao ;
Liu, Chao .
COMMUNICATIONS IN NONLINEAR SCIENCE AND NUMERICAL SIMULATION, 2012, 17 (02) :1036-1042
[6]   Day-Ahead Deregulated Electricity Market Price Forecasting Using Recurrent Neural Network [J].
Anbazhagan, S. ;
Kumarappan, N. .
IEEE SYSTEMS JOURNAL, 2013, 7 (04) :866-872
[7]  
[Anonymous], 2014, ADV NEURAL INFORM PR
[8]  
[Anonymous], 2018, ARXIV181010863
[9]  
[Anonymous], 2016, ARXIV161101779
[10]  
[Anonymous], 2017, ARXIV170304730