Skip to main content

Advertisement

Springer Nature Link
Log in

Design of an integrated model combining recurrent convolutions and attention mechanism for time series prediction

  • Published:
The Journal of Supercomputing Aims and scope Submit manuscript

Abstract

In applications such as healthcare, finance, and environmental monitoring, the demand for more reliable time-series prediction models has grown critical. Traditional models, such as VARMAx, struggle with capturing non–linear and complex dependencies inherent in sequential data. To address these challenges, this work proposes a hybrid model combining Long Short-Term Memory (LSTM) networks with Convolutional Neural Networks (CNNs) and incorporating attention mechanisms for improved precision and interpretability. LSTM networks are utilized to capture long-term dependencies in sequential data, while CNNs are employed to extract significant local features. The attention mechanism enhances the model’s focus on critical time-series instances, improving prediction accuracy and interpretability. Additionally, hyperparameter optimization is achieved using metaheuristic approaches such as the grey wolf optimizer and the coot optimization algorithm, ensuring maximum performance. The model integrates multimodal LSTMs to handle diverse data types, such as text and images, while preserving relationships between entities using Graph Neural Networks (GNNs). Adaptive feedback learning, combining reinforcement and federated learning, allows for real-time model adaptability while maintaining data privacy. Bayesian neural networks with dropout regularization provide uncertainty estimation, delivering confidence intervals alongside predictions. The proposed hybrid model demonstrates a 6–8% absolute improvement in predictive accuracy, reduced RMSE, and enhanced interpretability compared to traditional benchmarks. Its effectiveness is particularly evident in scenarios with high uncertainty, complex data, and the need for real-time model adaptation, setting a new standard in time-series prediction.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7

Similar content being viewed by others

Data availability

Kaggle and UCI machine learning repository platform. No datasets were generated or analysed during the current study.

Code availability

The code that supports the findings of this study is openly available at the following https://github.com/yuvaraja2417/RCNN--LSTM-with-GWO.

Abbreviations

CNN:

Convolutional Neural Network

LSTM:

Long Short-Term Memory

BiLSTM:

Bidirectional Long Short-Term Memory

GWO:

Grey Wolf Optimizer

COA:

Coot Optimization Algorithm

BLS:

Broad Learning System

MCMC:

Markov Chain Monte Carlo

MUTS:

Multivariate Utility Time-Series

IoT:

Internet of Things

GrC:

Granular Computing

SSA:

Salp Swarm Algorithm

AutoML:

Automated Machine Learning

QoS:

Quality of Service

PSO:

Particle Swarm Optimization

ESN:

Echo State Network

FCM:

Fuzzy Cognitive Map

DTW:

Dynamic Time Warping

FIG:

Fuzzy Information Granule

SDTW:

Standard Dynamic Time Warping

VAR:

Vector Autoregression

RNN:

Recurrent Neural Network

STD:

Seasonal-Trend-Dispersion

TCN:

Time Convolutional Network

MTL:

Multitask Learning

CPS:

Cyber-Physical Systems

GrM:

Granular Model

LSP:

Latent Time Graph Neural Network

References

  1. Kim D-K, Kim K (2022) A convolutional transformer model for multivariate time series prediction. IEEE Access 10:101319–101329

    Article  MATH  Google Scholar 

  2. Feng S et al (2020) Learning both dynamic-shared and dynamic-specific patterns for chaotic time-series prediction. IEEE Trans Cybern 52(6):4115–4125

    Article  MATH  Google Scholar 

  3. Zhou X et al (2022) Time series prediction method of industrial process with limited data based on transfer learning. IEEE Trans Ind Inform 19(5):6872–6882

    Article  MATH  Google Scholar 

  4. Yi J et al (2022) Intergroup cascade broad learning system with optimized parameters for chaotic time series prediction. IEEE Trans Artif Intell 3(5):709–721

    Article  MATH  Google Scholar 

  5. Chen X, Sun L (2021) Bayesian temporal factorization for multidimensional time series prediction. IEEE Trans Pattern Anal Mach Intell 44(9):4659–4673

    MATH  Google Scholar 

  6. Mubang F, Hall LO (2022) Vam: an end-to-end simulator for time series regression and temporal link prediction in social media networks. IEEE Trans Comput Social Syst 10(4):1479–1490

    Article  MATH  Google Scholar 

  7. Ma C, Dai G, Zhou J (2021) Short-term traffic flow prediction for urban road sections based on time series analysis and LSTM_BILSTM method. IEEE Trans Intell Transp Syst 23(6):5615–5624

    Article  MATH  Google Scholar 

  8. Ren S et al (2022) Coupledmuts: Coupled multivariate utility time-series representation and prediction. IEEE Internet Things J 9(22):22972–22982

    Article  Google Scholar 

  9. Yang Z et al (2020) Adaptive temporal-frequency network for time-series forecasting. IEEE Trans Knowl Data Eng 34(4):1576–1587

    MATH  Google Scholar 

  10. Na X et al (2020) Modified BBO-based multivariate time-series prediction system with feature subset selection and model parameter optimization. IEEE Trans Cybern 52(4):2163–2173

    Article  MATH  Google Scholar 

  11. Ma C et al (2022) The long-term prediction of time series: a granular computing-based design approach. IEEE Trans Syst, Man, and Cybern: Syst 52(10):6326–6338

    Article  MATH  Google Scholar 

  12. Zhou J et al (2023) Spatial context-aware time-series forecasting for QoS prediction. IEEE Trans Netw Service Manag 20(2):918–931

    Article  MATH  Google Scholar 

  13. Hu Y, Xiao F (2022) Time-series forecasting based on fuzzy cognitive visibility graph and weighted multisubgraph similarity. IEEE Trans Fuzzy Syst 31(4):1281–1293

    Article  MATH  Google Scholar 

  14. Yao Y et al (2022) Multivariate time-series prediction in industrial processes via a deep hybrid network under data uncertainty. IEEE Trans Ind Inform 19(2):1977–1987

    Article  MATH  Google Scholar 

  15. Boddu Y et al (2024) Design of an iterative dual metaheuristic VARMAx model enhancing efficiency of time series predictions. IEEE Access. https://doi.org/10.1109/ACCESS.2024.3454540

    Article  MATH  Google Scholar 

  16. Zhu C et al (2023) Long-term time series forecasting with multi-linear trend fuzzy information granules for LSTM in a periodic framework. IEEE Trans Fuzzy Syst 32(1):322–336

    Article  MATH  Google Scholar 

  17. Met I, Erkoç A, Seker SE (2022) Performance, efficiency, and target setting for bank branches: time series with automated machine learning. IEEE Access. 11:1000–1010

    Article  MATH  Google Scholar 

  18. Puri C et al (2022) Forecasting time series in healthcare with Gaussian processes and dynamic time warping based subset selection. IEEE J Biomed Health Inform 26(12):6126–6137

    Article  MATH  Google Scholar 

  19. Zheng W, Jun Hu (2022) Multivariate time series prediction based on temporal change information learning method. IEEE Trans Neural Netw Learn Syst 34(10):7034–7048

    Article  MATH  Google Scholar 

  20. Yan J et al (2022) Large-area land-cover changes monitoring with time-series remote sensing images using transferable deep models. IEEE Trans Geosci Remote Sensing 60:1–17

    Google Scholar 

  21. Ren W, Ma D, Han M (2022) Multivariate time series predictor with parameter optimization and feature selection based on modified binary salp swarm algorithm. IEEE Trans Industr Inf 19(4):6150–6159

    Article  MATH  Google Scholar 

  22. Wang Z et al (2023) Representing multi-view time-series graph structures for multivariate long-term time-series forecasting. IEEE Trans Artif Intell. 5(6):2651

    Article  MATH  Google Scholar 

  23. Koç E, Koç A (2022) Fractional fourier transform in time series prediction. IEEE Signal Process Lett 29:2542–2546

    Article  MATH  Google Scholar 

  24. Dudek G (2023) STD: a seasonal-trend-dispersion decomposition of time series. IEEE Trans Knowl Data Eng 35(10):10339–10350

    Article  MATH  Google Scholar 

  25. Li, Lina, et al. “Online Attention Enhanced Differential and Decomposed LSTM for Time Series Prediction.” IEEE Access (2024).

  26. Huang J et al (2022) Error-driven chained multiple-subnetwork echo state network for time-series prediction. IEEE Sensors J 22(20):19533–19542

    Article  Google Scholar 

  27. Pan J et al (2022) DUMA: dual mask for multivariate time series anomaly detection. IEEE Sensors J 23(3):2433–2442

    Article  MATH  Google Scholar 

  28. Zhan J et al (2024) A fuzzy C-means clustering-based hybrid multivariate time series prediction framework with feature selection. IEEE Trans Fuzzy Syst 32:4270–4284

    Article  MATH  Google Scholar 

  29. Na X et al (2022) Hierarchical echo state network with sparse learning: a method for multidimensional chaotic time series prediction. IEEE Trans Neural Netw Learn Syst 34(11):9302–9313

    Article  MathSciNet  MATH  Google Scholar 

  30. Zhao X et al (2022) Real-time multistep time-series prediction of driver’s head pose during IVIS secondary tasks for Human-Machine codriving and distraction warning systems. IEEE Sensors J 22(24):24364–24379

    Article  Google Scholar 

  31. Wu P et al (2024) A low-rank bayesian temporal matrix factorization for the transfer time prediction between metro and bus systems. IEEE Trans Intell Transp Syst 25(7):7206

    Article  MATH  Google Scholar 

  32. Fanjiang Y-Y, Syu Y, Huang W-L (2020) Time series QoS forecasting for Web services using multi-predictor-based genetic programming. IEEE Trans Serv Comput 15(3):1423–1435

    Article  MATH  Google Scholar 

  33. Zhou Z et al (2024) Kalman-SSM: modeling long-term time series with kalman filter structured state spaces. IEEE Signal Process Lett 31:2470

    Article  MATH  Google Scholar 

  34. Jastrzebska A et al (2021) Fuzzy cognitive map-driven comprehensive time-series classification. IEEE Trans Cybern 53(2):1348–1359

    Article  MATH  Google Scholar 

  35. Martínez F et al (2022) Time series forecasting by generalized regression neural networks trained with multiple series. IEEE Access 10:3275–3283

    Article  MATH  Google Scholar 

  36. Harerimana G, Kim JW, Jang B (2022) A multi-headed transformer approach for predicting the patient’s clinical time-series variables from charted vital signs. IEEE Access 10:105993–106004

    Article  MATH  Google Scholar 

  37. Guo J et al (2022) Structure parameter optimized kernel based online prediction with a generalized optimization strategy for nonstationary time series. IEEE Trans Signal Process 70:2698–2712

    Article  MathSciNet  MATH  Google Scholar 

  38. Wang Y et al (2022) The trend-fuzzy-granulation-based adaptive fuzzy cognitive map for long-term time series forecasting. IEEE Trans Fuzzy Syst 30(12):5166–5180

    Article  MATH  Google Scholar 

  39. Pranolo A et al (2022) Robust LSTM With tuned-PSO and bifold-attention mechanism for analyzing multivariate time-series. Ieee Access 10:78423–78434

    Article  Google Scholar 

  40. Li L et al (2024) A recurrent spatio-temporal graph neural network based on latent time graph for multi-channel time series forecasting. IEEE Signal Process Lett 31:2875

    Article  MATH  Google Scholar 

  41. Lin Y (2024) Long-term traffic flow prediction using stochastic configuration networks for smart cities. IECE Trans Intell System 1(2):79–90

    Article  MATH  Google Scholar 

  42. Ahmed RS et al (2024) Comparison of deep learning algorithms for retail sales forecasting. IECE Trans Intell Syst 1(3):112–126

    Article  MATH  Google Scholar 

  43. Shi H et al (2024) Adaptive tunable predefined-time backstepping control for uncertain robotic manipulators. IECE Trans Sensing, Commun Control 1(2):126–135

    Article  MATH  Google Scholar 

  44. Jin X-B et al (2025) Ada-STGMAT: An adaptive spatio-temporal graph multi-attention network for intelligent time series forecasting in smart cities. Expert Syst Appl 269:126428

    Article  MATH  Google Scholar 

Download references

Funding

The authors declare that no funding was received for the conduct of this research or the preparation of this manuscript.

Author information

Authors and Affiliations

  1. Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation, Bowrampet, Hyderabad, Telangana, 500043, India

    Nirmalajyothi Narisetty

  2. Department of Computer Science and Engineering, Narasaraopeta Engineering College (Autonomous), Narasaraopet, Palnadu(Dist), Andhra Pradesh, 522601, India

    Kunda Suresh Babu

  3. Department of Computer Science and Engineering (AI&ML), Malla Reddy College of Engineering & Technology, Hyderabad, Telangana, 500100, India

    Lakshmi Naga Jayaprada Gavarraju

  4. Department of Computer Science and Engineering, Malla Reddy University, Hyderabad, Telangana, 500100, India

    Munigeti Benjmin Jashva

  5. Department of Computer Science and Engineering (AI&ML), Bhoj Reddy Engineering College for Women, Vinay Nagar, Saidabad, Hyderabad, Telangana, 500059, India

    Seshu Bhavani Mallampati

  6. Department of Mathematics, SAS, VIT -AP University, Amaravati, Andhra Pradesh, 522241, India

    Yuvaraja Boddu

Authors
  1. Nirmalajyothi Narisetty
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Kunda Suresh Babu
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Lakshmi Naga Jayaprada Gavarraju
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Munigeti Benjmin Jashva
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Seshu Bhavani Mallampati
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Yuvaraja Boddu
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

NirmalajyothiNarisetty contributed to the methodology and overall study design. Kunda Suresh Babu was responsible for data analysis and interpretation. Lakshmi Naga Jayaprada Gavarraju assisted in the literature review and manuscript writing. MunigetiBenjmin Jashva contributed to the experimental design and implementation. Seshu Bhavani Mallampati provided support in data collection and preprocessing. Yuvaraja Boddu led the conceptualization and implementation of the project, ensuring its integrity and coherence.

Corresponding author

Correspondence to Kunda Suresh Babu.

Ethics declarations

Conflicts of interest

The authors declare no competing interests.

Human or Animal Rights

Not applicable.

Informed consent

Not Applicable.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Narisetty, N., Babu, K.S., Gavarraju, L.N.J. et al. Design of an integrated model combining recurrent convolutions and attention mechanism for time series prediction. J Supercomput 81, 642 (2025). https://doi.org/10.1007/s11227-025-07154-5

Download citation

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s11227-025-07154-5

Keywords