Skip to main content

Advertisement

Springer Nature Link
Log in

A graph neural network-based stock forecasting method utilizing multi-source heterogeneous data fusion

  • Published:
Multimedia Tools and Applications Aims and scope Submit manuscript

Abstract

The study of the prediction of stock market volatility is of great significance to rationally control financial market risks and increase excessive investment returns and has received extensive attention from academic and commercial circles. However, as a dynamic and complex system, the stock market is affected by multiple factors and has a comprehensive capability to include complex financial data. Given that the explanatory variables of influencing factors are diverse, heterogeneous and complex, the existing intelligent algorithms have great limitations for the analysis and processing of multi-source heterogeneous data in the stock market. Therefore, this study adopts the edge weight and information transmission mechanism suitable for subgraph data to complete node screening, the gate recurrent unit (GRU) and long short-term memory (LSTM) to aggregate subgraph nodes. The compiled data contain the metapaths of three types of index data, and the introduction of the association relationship attention dimension effectively mines the implicit meanings of multi-source heterogeneous data. The metapath attention mechanism is combined with a graph neural network to complete the classification of multi-source heterogeneous graph data, by which the prediction of stock market volatility is realized. The results show that the above method is feasible for the fusion of heterogeneous stock market data and the mining of implicit semantic information of association relations. The accuracy of the proposed method for the prediction of stock market volatility in this study is 16.64% higher than that of the dimensional reduction index and 14.48% higher than that of other methods for the fusion and prediction of heterogeneous data using the same model.

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
Fig. 8
Fig. 9
Fig. 10

Similar content being viewed by others

Explore related subjects

Discover the latest articles and news from researchers in related subjects, suggested using machine learning.

Notes

  1. https://pan.baidu.com/s/1Wlj7FewoUDcwMWBET58VNg.

References

  1. Arasu A, Widom J. Resource sharing in continuous sliding-window aggregates[EB/OL]. [2019-10-02]. https://www.microsoft.com/en-us/research/wp-content/uploads/2016/02/sharing.pdf

  2. Atkins A, Niranjan M, Gerding E (2018) Financial news predicts stock market volatility better than close price[J]. J Financ Data Sci 4(2):120–137

    Article  Google Scholar 

  3. Belov G, Scheithauer G (2006) A branch-and-cut-and-price algorithm for one-dimensional stock cutting and two-dimensional two-stage cutting[J]. Eur J Oper Res 171(1):85–106

    Article  MathSciNet  MATH  Google Scholar 

  4. Box GEP, Jenkins GM, Reinsel GC et al (2015) Time series analysis: forecasting and control[M]. Wiley, Hoboken

  5. Bruna J, Zaremba W, Szlam A et al (2013) Spectral networks and locally connected networks on graphs[J]. arXiv preprint arXiv:1312.6203

  6. Bulkowski TN (2012) Encyclopedia of Canlestick charts[M]. Wiley, Hoboken

  7. Chai L, Xu H, Luo Z et al (2020) A multi-source heterogeneous data analytic method for future price fluctuation prediction[J]. Neurocomputing 418:11–20

    Article  Google Scholar 

  8. Chan WS (2003) Stock price reaction to news and no-news: drift and reversal after headlines[J]. J Financial Econ 70(2):223–260

    Article  Google Scholar 

  9. Chen Y, Hao Y (2017) A feature weighted support vector machine and k-nearest neighbor algorithm for stock market indices prediction [J]. Expert Syst Appl 80:340–355

    Article  Google Scholar 

  10. De Gooijer JG, Hyndman RJ (2006) 25 years of time series forecasting[J]. Int J Forecast 22(3):443–473

    Article  Google Scholar 

  11. Defferrard M, Bresson X, Vandergheynst P (2016) Convolutional neural networks on graphs with fast localized spectral filtering[C]. In: Proceedings of Advances in Neural Information Processing Systems, 3844–3852

  12. Ding X, Zhang Y, Liu T et al (2014) Using structured events to predict stock price movement: An empirical investigation[C]. In: Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), 1415–1425

  13. Ding X, Zhang Y, Liu T et al (2015) Deep learning for event-driven stock prediction[C]. In: Proceedings of the Twenty-fourth International Joint Conference on Artificial Intelligence

  14. Edwards RD, Magee J, Bassetti WHC (2018) Technical analysis of stock trends[M]. CRC Press, Boca Raton

  15. Fama EF (1970) Efficient capital markets: A review of theory and empirical work[J]. J Finance 25(2):383–417

    Article  Google Scholar 

  16. Fama EF, French KR (1992) The cross-section of expected stock returns[J]. J Financ 47(2):427–465

    Article  Google Scholar 

  17. Fama EF, French KR (1993) Common risk factors in the returns on stocks and bonds[J]. J Financ Econ 33(1):3–56

    Article  MATH  Google Scholar 

  18. French FKR (1996) Multifactor explanations of asset pricing anomalies[J]. J Financ 51(1):55–84

    Article  Google Scholar 

  19. Goodfellow I, Bengio Y, Courville A et al (2016) Deep learning[M]. MIT Press, Cambridge

    MATH  Google Scholar 

  20. Granville JE (1960) A strategy of daily stock market timing for maximum profit[M]. Prentice-Hall, Hoboken

  21. Guo J-Y, Li R-H(2020) Graph neural network based anomaly detection in dynamic networks[J]. J Softw 31(03):156–170

    Google Scholar 

  22. Huang TL (2018) The puzzling media effect in the Chinese stock market[J]. Pac-Basin Financ J 49:129–146

    Article  Google Scholar 

  23. Jegadeesh N, Titman S (1993) Returns to buying winners and selling losers: Implications for stock market efficiency[J]. J Financ 48(1):65–91

    Article  Google Scholar 

  24. Jiao G, Zhang Y(2019) Research on user participation behavior of online stock community[J]. J Inf Syst 1

  25. Kahneman D (2003) Maps of bounded rationality: Psychology for behavioral economics[J]. Am Econ Rev 93(5):1449–1475

    Article  Google Scholar 

  26. Kim R, So CH, Jeong M et al. (2019) Hats: A hierarchical graph attention network for stock movement prediction[J]. arXiv preprint arXiv:1908.07999

  27. Kipf TN, Welling M (2016)Semi-supervised classification with graph convolutional networks[J]. arXiv preprint arXiv:1609.02907

  28. Kusuma RM, I, Ho TT, Kao WC et al (2019) Using deep learning neural networks and candlestick chart representation to predict stock market[J]. arXiv preprint arXiv:1903.12258

  29. Li Y, Tarlow D, Brockschmidt M et al (2015) Gated graph sequence neural networks[J]. arXiv preprint arXiv:1511.05493

  30. Li Q, Jiang LL, Li P et al (2015)Tensor-based learning for predicting stock movements[C]. In: Proceedings of the Twenty-ninth AAAI Conference on Artificial Intelligence

  31. Li Q, Wang J, Wang F et al (2017) The role of social sentiment in stock markets: a view from joint effects of multiple information sources[J]. Multimed Tools Appl 76(10):12315–12345

    Article  Google Scholar 

  32. Li Lihui T, Xiang Y, Haidong et al (2005) Financial time series forecasting based on SVR[J]. Comput Eng Appl 41(30):221–224

    Google Scholar 

  33. Liu X, Dou Y, Yin J et al (2016) Multiple kernel k-means clustering with matrix-induced regularization[C]. In: Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence, 1888–1894

  34. Liu Y, Zeng Q, Yang H et al (2018) Stock price movement prediction from financial news with deep learning and knowledge graph embedding[C]. In: Proceedings of the Pacific Rim Knowledge Acquisition Workshop. Springer, Cham, 102–113

  35. Liu J, Lu Z, Du W (2019) Combining enterprise knowledge graph and news sentiment analysis for stock price prediction[C]. In: Proceedings of the 52nd Hawaii International Conference on System Sciences

  36. Lo AW, MacKinlay AC (1988) Stock market prices do not follow random walks: Evidence from a simple specification test[J]. Rev Financ Stud 1(1):41–66

    Article  Google Scholar 

  37. Matsunaga D, Suzumura T, Takahashi T (2019) Exploring graph neural networks for stock market predictions with rolling window analysis[J]. arXiv preprint arXiv:1909.10660

  38. Menon VK, Vasireddy NC, Jami SA et al (2016) Bulk price forecasting using spark over nse data set[C]. In: Proceedings of International Conference on Data Mining and Big Data. Springer, Cham, 137–146

  39. Mittal A, Goel A (2012) Stock prediction using twitter sentiment analysis[J]. Standford University, 15

  40. Qu Q, Yu H, Huang R (2018) Spammer detection technology of social network based on graph convolution network[J]. J Netw Inform Secur 004(005):39–46

    Google Scholar 

  41. Rojas I, Valenzuela O, Rojas F et al (2008)Soft-computing techniques and ARMA model for time series prediction[J]. Neurocomputing 71(4–6):519–537

    Article  Google Scholar 

  42. Roondiwala M, Patel H, Varma S (2017) Predicting stock prices using LSTM[J]. Int J Sci Res (IJSR) 6(4):1754–1756

    Google Scholar 

  43. Shihavuddin A, Ambia MN, Arefin M et al (2010) Prediction of stock price analyzing the online financial news using Naive Bayes classifier and local economic trends [C]. In: Proceedings of the 3rd International Conference on Advanced Computer Theory and Engineering. Piscataway, IEEE, 22–26

  44. Shiller RJ (2015) Irrational exuberance: Revised and expanded third edition[M]. Princeton University Press, Princeton

  45. Si J, Mukherjee A, Liu B et al (2013) Exploiting topic based twitter sentiment for stock prediction[C]. In: Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics, vol 2: Short Papers, 24–29

  46. Simon HA (1996) Designing organizations for an information-rich world[J]. Int Libr Crit Writ Econ 70:187–202

    Google Scholar 

  47. Tan J, Wang J, Rinprasertmeechai D et al (2019) A Tensor-based eLSTM model to predict stock price using financial news[C]. In: Proceedings of the 52nd Hawaii International Conference on System Sciences

  48. Tanaka-Yamawaki M, Tokuoka S (2007) Adaptive use of technical indicators for the prediction of intra-day stock prices[J]. Phys A 383(1):125–133

    Article  Google Scholar 

  49. Tang H, Chiu KC, Xu L (2003) Finite mixture of ARMA-GARCH model for stock price prediction[C]. In: Proceedings of the Third International Workshop on Computational Intelligence in Economics and Finance (CIEF’2003), North Carolina, USA, 1112–1119

  50. Tsai CF, Quan ZY (2014) Stock prediction by searching for similarities in Canlestick charts[J]. ACM Trans Manage Inform Syst (TMIS) 5(2):9

    Google Scholar 

  51. Wei YC, Lu YC, Chen JN et al (2017) Informativeness of the market news sentiment in the Taiwan stock market[J]. North Am J Econ Financ 39:158–181

    Article  Google Scholar 

  52. Zhang X, Li Y, Wang S et al (2018) Enhancing stock market prediction with extended coupled hidden markov model over multi-sourced data[J]. Knowl Inf Syst

  53. Zhang X, Zhang Y, Wang S et al (2018) Improving stock market prediction via heterogeneous information fusion[J]. Knowl Based Syst 143:236–247

    Article  Google Scholar 

Download references

Acknowledgements

The research work is supported by the National Natural Science Foundation of China (NSFC) (71873108 and 62072379), Fundamental Research Funds for the Central Universities kjcx20210103), Financial Intelligence and Financial Engineering Key Lab of Sichuan Province, Jiaozi Institute of Financial Technology Innovation, Southwest University of Finance and Economics (cgzh20210204), Research Program of Science and Technology at Universities of Inner Mongolia Autonomous Region (2021GG0164) and Financial Innovation Center of the Southwestern University of Finance and Economics.

Author information

Authors and Affiliations

  1. School of Economic Information Engineering, Southwestern University of Finance and Economics, 610000, Cheng Du, China

    Xiaohan Li, Jun Wang, Jinghua Tan, Shiyu Ji & Huading Jia

Authors
  1. Xiaohan Li
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Jun Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Jinghua Tan
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Shiyu Ji
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Huading Jia
    View author publications

    You can also search for this author inPubMed Google Scholar

Corresponding author

Correspondence to Xiaohan Li.

Additional information

Publisher’s note

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

Appendix 1

Appendix 1

1.1 Trading subgraph data construction

The trading subgraph data construction process is illustrated in the following table. First, the program defines the interval of the determined trading day and builds a subgraph DGLGraph(). Then, based on the trading day, subgraph nodes are set up with the add_nodes() function. Finally, the trading day data index is taken as the node feature via ndata().

figure a

Subgraph data construction for stock market news

To construct the stock market news subgraph, two core functions are used. The first function, EWEIGHT(), is mainly used to calculate the similarities between news events and the edge weight values for the construction of the subgraph data. Jieba () is used for word classification, and SparseMatrixSimilarity () is used to calculate the similarity between texts. The second function is CREATENEWSUB (). It acts as a subgraph builder function, where each piece of news text is used as a node, Add_edge () increases the edges between the nodes, and EWEIGHT() uses the edge weights to assign values. The news text vector contains node characteristics.

figure b

Construction of graphical index subgraph data

Two core functions are applied to construct graphical index subgraph data. The first function, LOCATE(), is mainly used to obtain the position of a number in a vector. The second function, GRAPHICSUB(), is used as the construction function for the graphical index subgraph. A ‘for’ loop and cv2.split are used to extract the three primary color features of the graphical index, and the characteristic value is used as the corresponding node feature. Then, the second ‘for’ loop of the function is used to establish edges between the characteristic value nodes and the characteristic value position nodes.

figure c

The shared link details the procedures in this articleFootnote 1

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Li, X., Wang, J., Tan, J. et al. A graph neural network-based stock forecasting method utilizing multi-source heterogeneous data fusion. Multimed Tools Appl 81, 43753–43775 (2022). https://doi.org/10.1007/s11042-022-13231-1

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11042-022-13231-1

Keywords