Cyclic Gate Recurrent Neural Networks for Time Series Data with Missing Values

P.B. Weerakody; K.W. Wong; G. Wang

doi:10.1007/s11063-022-10950-2

Back

Cyclic Gate Recurrent Neural Networks for Time Series Data with Missing Values

Journal article

Open access

Peer reviewed

Cyclic Gate Recurrent Neural Networks for Time Series Data with Missing Values

P.B. Weerakody, K.W. Wong and G. Wang

Neural processing letters, Vol.55(2), pp.1527-1554

01/04/2023

DOI: https://doi.org/10.1007/s11063-022-10950-2

Appears in Open Access via Read & Publish Agreements

Files and links (2)

pdf

Cyclic Gate Recurrent Neural Networks for Time Series Data with Missing Values2.06 MBDownload View

Published (Version of Record)CC BY V4.0, Open Access

url

https://doi.org/10.1007/s11063-022-10950-2View

Published (Version of Record) Open

Abstract

Article

Artificial Intelligence

Complex Systems

Computational Intelligence

Computer Science

Gated Recurrent Neural Networks (RNNs) such as LSTM and GRU have been highly effective in handling sequential time series data in recent years. Although Gated RNNs have an inherent ability to learn complex temporal dynamics, there is potential for further enhancement by enabling these deep learning networks to directly use time information to recognise time-dependent patterns in data and identify important segments of time. Synonymous with time series data in real-world applications are missing values, which often reduce a model’s ability to perform predictive tasks. Historically, missing values have been handled by simple or complex imputation techniques as well as machine learning models, which manage the missing values in the prediction layers. However, these methods do not attempt to identify the significance of data segments and therefore are susceptible to poor imputation values or model degradation from high missing value rates. This paper develops Cyclic Gate enhanced recurrent neural networks with learnt waveform parameters to automatically identify important data segments within a time series and neglect unimportant segments. By using the proposed networks, the negative impact of missing data on model performance is mitigated through the addition of customised cyclic opening and closing gate operations. Cyclic Gate Recurrent Neural Networks are tested on several sequential time series datasets for classification performance. For long sequence datasets with high rates of missing values, Cyclic Gate enhanced RNN models achieve higher performance metrics than standard gated recurrent neural network models, conventional non-neural network machine learning algorithms and current state of the art RNN cell variants.

Details

Title: Cyclic Gate Recurrent Neural Networks for Time Series Data with Missing Values
Authors/Creators: P.B. Weerakody (Author/Creator) - Murdoch University
K.W. Wong (Author/Creator)
G. Wang (Author/Creator)
Publication Details: Neural processing letters, Vol.55(2), pp.1527-1554
Publisher: Springer US
Grant note: Murdoch University
Identifiers: 991005544918607891
Murdoch Affiliation: Murdoch University
Language: English
Resource Type: Journal article

Metrics

28 File views/ downloads

223 Record Views

24 Times Cited - Web of Science

InCites Highlights

These are selected metrics from InCites Benchmarking & Analytics tool, related to this output

Citation topics: 1 Clinical & Life Sciences; 1.155 Medical Ethics; 1.155.2774 Artificial Intelligence in Healthcare and Medicine
Web Of Science research areas: Computer Science, Artificial Intelligence
ESI research areas: Computer Science