Blog
a-mf-convlstm-xai-model-integrating-multi-feature-and-fuzzy-control-for-financial-time-series-forecasting-–-scientific-reports

A MF-ConvLSTM-XAI model integrating multi-feature and fuzzy control for financial time series forecasting – Scientific Reports

  • Article
  • Open access
  • Published:
  • Ruimin Liu1,
  • Shuihan Yi2,
  • Arman Ablikim3 &
  • …
  • Xiaonan Song4 

Scientific Reports , Article number:  (2025) Cite this article

We are providing an unedited version of this manuscript to give early access to its findings. Before final publication, the manuscript will undergo further editing. Please note there may be errors present which affect the content, and all legal disclaimers apply.

Subjects

Abstract

In financial time series prediction, existing methods struggle to fully capture the dynamic characteristics and nonlinear relationships of the data. Furthermore, they lack sufficient capability to integrate multiple features (such as prices, trading volumes, and technical indicators), particularly in highly volatile market conditions, where the prediction accuracy and stability of these models tend to be poor. This study proposes a financial time series forecasting (FTSF) model named Multi-Feature Convolutional Long Short-Term Memory Network–Explainable Artificial Intelligence (MF-ConvLSTM-XAI). By deeply integrating Convolutional Long Short-Term Memory Networks (ConvLSTM) with Explainable Artificial Intelligence (XAI) techniques, the model improves the accuracy and stability of financial time series predictions and enhances interpretability. Firstly, the sliding window technique is applied to segment the original financial time series, with data in each window transformed into multiple feature representations. The features analyzed encompass price fluctuations, trading volumes, and technical indicators, enabling the model to identify intricate patterns within the series. Subsequently, the Gramian Angular Difference Fields (GADF) technique is employed to transform the extracted sequential features into image formats. Regarding model design, the MF-ConvLSTM network leverages both convolutional neural networks (CNNs) and Long Short-Term Memory (LSTM) networks. Additionally, a Fuzzy Control Mechanism (FCM) is integrated into the input gate of the ConvLSTM to facilitate subspace reconstruction, enhancing the information flow and state update process in the memory units. To enhance the model’s interpretability, XAI technology is applied to feature importance analysis, decision path visualization, and model explanation reports. This helps researchers and practitioners better understand the model’s decision-making process, increasing the model’s transparency and credibility. To evaluate the effectiveness of the MF-ConvLSTM-XAI model, experiments are conducted using the New York Stock Exchange Composite Index dataset. The results indicate that MF-ConvLSTM-XAI achieves higher prediction accuracy compared with traditional LSTM models and other baseline methods. Specifically, the Mean Squared Error (MSE) on the test set decreases by approximately 15%, and the Mean Absolute Error (MAE) decreases by about 12%. Moreover, the model demonstrates greater robustness under highly volatile market conditions. These findings validate the effectiveness and advantages of MF-ConvLSTM-XAI in FTSF. Overall, by integrating multi-feature information and Fuzzy Cognitive Mechanisms (FCM), MF-ConvLSTM-XAI shows a positive impact on improving the accuracy and stability of financial time series predictions. Although the study achieves promising results on the New York Stock Exchange dataset, the model’s generalization capability still requires further validation across more diverse financial datasets, including cryptocurrency and emerging markets. Nonetheless, this approach provides a valuable tool for financial data analysis and offers new insights for research and applications in the field.

Data availability

Data is provided within the manuscript or supplementary information files.

References

  1. Yolcu, O. C. & Yolcu, U. A novel intuitionistic fuzzy time series prediction model with cascaded structure for financial time series. Expert Syst. Appl. 215, 119336 (2023).

    Google Scholar 

  2. Gajamannage, K., Park, Y. & Jayathilake, D. I. Real-time forecasting of time series in financial markets using sequentially trained dual-LSTMs. Expert Syst. Appl. 223, 119879 (2023).

    Google Scholar 

  3. Lazcano, P. J., Herrera & Monge, M. A combined model based on recurrent neural networks and graph convolutional networks for financial time series forecasting. Math 11, 224 (2023).

    Google Scholar 

  4. Lee, H. Y., Beh, W. L. & Lem, K. H. Forecasting with information extracted from the residuals of ARIMA in financial time series using continuous wavelet transform. Int. J. Bus. Intell. Data Min. 22, 70–99 (2023).

    Google Scholar 

  5. Kurani, P. et al. A comprehensive comparative study of artificial neural network (ANN) and support vector machines (SVM) on stock forecasting. Ann. Data Sci. 10, 183–208 (2023).

    Google Scholar 

  6. Chen, W. J., Yao, J. J. & Shao, Y. H. Volatility forecasting using deep neural network with time-series feature embedding. Ekonomska istraživanja. 36, 1377–1401 (2023).

    Google Scholar 

  7. Usmani, S. & Shamsi, J. A. LSTM based stock prediction using weighted and categorized financial news. PLoS ONE. 18, e0282234 (2023).

    Google Scholar 

  8. Vuletić, M., Prenzel, F. & Cucuringu, M. Fin-gan: forecasting and classifying financial time series via generative adversarial networks. Quant. Finance. 24, 175–199 (2024).

    Google Scholar 

  9. Zhang, N. N. A., Sjarif & Ibrahim, R. Deep learning models for price forecasting of financial time series: A review of recent advancements: 2020–2022. Wiley Interdiscip Rev. Data Min. Knowl. Discov. 14, e1519 (2024).

    Google Scholar 

  10. Cheng, F. et al. Financial time series forecasting with multi-modality graph neural network. Pattern Recognit. 121, 108218 (2022).

    Google Scholar 

  11. Wu, K. et al. Price graphs: utilizing the structural information of financial time series for stock prediction. Inf. Sci. 588, 405–424 (2022).

    Google Scholar 

  12. Dudukcu, H. V. et al. Temporal convolutional networks with RNN approach for chaotic time series prediction. Appl. Soft Comput. 133, 109945 (2023).

    Google Scholar 

  13. Edalatpanah, S. A. et al. A hybrid time series forecasting method based on neutrosophic logic with applications in financial issues. Eng. Appl. Artif. Intell. 129, 107531 (2024).

    Google Scholar 

  14. Babii, E., Ghysels & Striaukas, J. Machine learning time series regressions with an application to nowcasting. J. Bus. Econ. Stat. 40, 1094–1106 (2022).

    Google Scholar 

  15. Masini, R. P., Medeiros, M. C. & Mendes, E. F. Machine learning advances for time series forecasting. J. Econ. Surv. 37, 76–111 (2023).

    Google Scholar 

  16. Zhang, Z., Zhou & Wu, R. Analyzing and predicting financial time series data using recurrent neural networks. J. Ind. Eng. Appl. Sci. 2, 15–21 (2024).

    Google Scholar 

  17. Song Analysis of china’s carbon market price fluctuation and international carbon credit financing mechanism using random forest model. PLoS ONE. 19, e0294269 (2024).

    Google Scholar 

  18. Wang, Y. & Yan, K. Machine learning-based quantitative trading strategies across different time intervals in the American market. Quant. Finance Econ. 7, 569–594 (2023).

    Google Scholar 

  19. Li, G. & Jung, J. J. Deep learning for anomaly detection in multivariate time series: Approaches, applications, and challenges. Inf. Fusion. 91, 93–102 (2023).

    Google Scholar 

  20. Benidis, K. et al. Deep learning for time series forecasting: tutorial and literature survey. ACM Comput. Surv. 55, 1–36 (2022).

    Google Scholar 

  21. He, K. et al. Financial time series forecasting with the deep learning ensemble model. Math 11, 1054 (2023).

    Google Scholar 

  22. Zaheer, S. et al. A multi parameter forecasting for stock time series data using LSTM and deep learning model. Math 11, 590 (2023).

    Google Scholar 

  23. Bello, H. O., Ige, A. B. & Ameyaw, M. N. Deep learning in high-frequency trading: conceptual challenges and solutions for real-time fraud detection. World J. Adv. Eng. Technol. Sci. 12, 35–46 (2024).

    Google Scholar 

  24. Theissler, F., Spinnato, U., Schlegel & Guidotti, R. Explainable AI for time series classification: a review, taxonomy and research directions. IEEE Access. 10, 100700–100724 (2022).

    Google Scholar 

  25. Çelik, T. B., İcan, Ö. & Bulut, E. Extending machine learning prediction capabilities by explainable AI in financial time series prediction. Appl. Soft Comput. 132, 109876 (2023).

    Google Scholar 

  26. Al-Najjar, H. A. et al. A novel method using explainable artificial intelligence (XAI)-based Shapley additive explanations for Spatial landslide prediction using time-series SAR dataset. Gondwana Res. 123, 107–124 (2023).

    Google Scholar 

  27. Li, S. et al. Prediction of the creeping of AFC based on fuzzy reasoning and Bi-LSTM fusion iteration. Artif. Intell. Rev. 57, 1–34 (2024).

    Google Scholar 

  28. Zhuang, Y. & Wei, H. Early warning model and prevention of regional financial risk integrated into legal system. PLoS ONE. 18, e0286685 (2023).

    Google Scholar 

  29. Júnior, J. S. S. et al. Survey on deep fuzzy systems in regression applications: A view on interpretability. Int. J. Fuzzy Syst. 25, 2568–2589 (2023).

    Google Scholar 

  30. Kan, N. H. L., Cao, Q. & Quek, C. Learning and processing framework using fuzzy deep neural network for trading and portfolio rebalancing. Appl. Soft Comput. 152, 111233 (2024).

    Google Scholar 

  31. Tan, J. C. M., Cao, Q. & Quek, C. FE-RNN: A fuzzy embedded recurrent neural network for improving interpretability of underlying neural network. Inf. Sci. 663, 120276 (2024).

    Google Scholar 

  32. Abril-Bermúdez, F. S. et al. Multifractality approach of a generalized Shannon index in financial time series. PLoS ONE. 19, e0303252 (2024).

    Google Scholar 

  33. Wu, J. et al. Imaging feature-based clustering of financial time series. PLoS ONE. 18, e0288836 (2023).

    Google Scholar 

  34. Fang, Z. et al. Movement forecasting of financial time series based on adaptive LSTM-BN network. Expert Syst. Appl. 213, 119207 (2023).

    Google Scholar 

  35. Tao, Z. et al. An integrated approach implementing sliding window and DTW distance for time series forecasting tasks. Appl. Intell. 53, 20614–20625 (2023).

    Google Scholar 

  36. Chen, L. et al. Classification and identification of extreme wind events by CNNs based on shapelets and improved GASF-GADF. J. Wind Eng. Ind. Aerodyn. 253, 105852 (2024).

    Google Scholar 

  37. Zhang, H., Wang, L. & Shi, W. Seismic control of adaptive variable stiffness intelligent structures using fuzzy control strategy combined with LSTM. J. Build. Eng. 78, 107549 (2023).

    Google Scholar 

  38. Abdelhady, S. et al. Green finance growth prediction model based on time-series conditional generative adversarial networks. PLoS ONE. 19, e0306874 (2024).

    Google Scholar 

  39. Tu, S. et al. Multi-step multidimensional statistical arbitrage prediction using PSO deep-ConvLSTM: an enhanced approach for forecasting price spreads. Appl. Sci. 14, 3798 (2024).

    Google Scholar 

  40. Duan & Ke, W. Advanced stock price prediction using LSTM and informer models. J. Artif. Intell. Gen. Sci. 5, 141–166 (2024).

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Université Paris-Dauphine, Paris, France

    Ruimin Liu

  2. Research Institute for Interdisciplinary Sciences, Shanghai University of Finance and Economics, Shanghai, China

    Shuihan Yi

  3. Beijing Borrui Data Technology Co., Ltd, Beijing, China

    Arman Ablikim

  4. MoxeAI, Inc, Xiaonan Song, Shenzhen, 100028, China

    Xiaonan Song

Authors

  1. Ruimin Liu
  2. Shuihan Yi
  3. Arman Ablikim
  4. Xiaonan Song

Contributions

Ruimin Liu and Shuihan Yi contributed to conception and design of the study. Arman Ablikim organized the database. Xiaonan Song performed the statistical analysis. Ruimin Liu wrote the first draft of the manuscript. Shuihan Yi, Arman Ablikim and Xiaonan Song wrote sections of the manuscript. All authors contributed to manuscript revision, read, and approved the submitted version.

Corresponding author

Correspondence to Xiaonan Song.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

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

Supplementary Information

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Liu, R., Yi, S., Ablikim, A. et al. A MF-ConvLSTM-XAI model integrating multi-feature and fuzzy control for financial time series forecasting. Sci Rep (2025). https://doi.org/10.1038/s41598-025-27121-7

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1038/s41598-025-27121-7

Keywords

colind88

Related Posts

unisys-named-a-global-leader-in-isg
Unisys Named a Global Leader in ISG

Company recognized for strategic excellence, experience-as-a-service offering, and innovative AI deployment , /PRNewswire/ -- Unisys (NYSE: UIS) has been named a global Leader by Information Services Group (ISG) in its 2025 Provider Lens® report for Generative AI (GenAI) Services. For the second consecutive year, Unisys earned recognition in two key report quadrants: Strategy and Consulting

Back To Top