1. NCCU Academic Hub
  2. Publications
  3. College of Commerce
  4. Theses
  5. Item

Publications-Theses

Article View/Open

Publication Export

Google ScholarTM

NCCU Library

Citation Infomation

Related Publications in TAIR

題名 圖神經網路於台灣股市長期股票排名預測之應用：基於橫截面數據與優化排序方法
Graph Neural Networks for Long-Term Stock Ranking Prediction in the Taiwan Stock Market: An Approach Based on Cross-Sectional Data and Optimized Learning to Rank
作者 鄭玉海
Cheng, Yu-Hai
貢獻者 黃泓智
鄭玉海
Cheng, Yu-Hai
關鍵詞 學習排序
圖神經網路
股票排名預測
LambdaRank
集成學習
量化投資
Learning to Rank
Graph Neural Network
Stock Ranking Prediction
LambdaRank
Ensemble Learning
Quantitative Investment
日期 2025
上傳時間 4-Aug-2025 14:11:01 (UTC+8)
摘要 圖神經網路（GNN）已經在多個領域展現出優良的表現，在股票排名預測中展現了捕捉複雜關聯的潛力，但當前主流模型普遍採用的「改良式 Pointwise 方法」存在理論缺陷，即「目標錯配」與「雜訊敏感」，導致模型在優化過程中易受金融市場低信噪比的影響。為解決此問題，本研究提出一個結合多層感知器（MLP）、圖注意力網路（GAT）與 Listwise 排序方法 LambdaRank 的新型股票排名預測框架。該框架首先利用 MLP 學 習個股的橫截面特徵，再透過 GAT 建模同產業股票間的關聯性，並以 LambdaRank 為優化目標，從根本上使模型的學習方向與排序任務的本質保持一致。此外，本研究設計並驗證了一套創新的「頂部排名優化策略」，引導模型聚焦於提升最具投資價值的頂級股票之預測準確性。 透過對台灣股市 2020 年至 2024 年的滾動窗口回測，本研究提出的完整模型展現了卓越的績效，顯著優於市場基準與基線模型。消融實驗進一步證實，「頂部排名優化」策略與「GNN 模塊」均為模型取得成功的關鍵組件，移除任何一者皆會導致績效顯著衰退。研究亦發現，透過與泛化能力更強的模型進行集成，能有效緩解本框架在較大選股數中可能存在的「局部過擬合」風險，從而構建出一個績效更佳且更穩健的投資決策系統。總體而言，本研究不僅提供了一個理論更健全、實證更有效的智能選股框架，也 為深度學習模型在金融市場的設計、評估與應用提供了深刻的洞見。
Graph Neural Networks (GNNs) have demonstrated superior performance across various domains and exhibit significant potential in capturing complex relationships for stock ranking prediction. However, prevailing GNN stock ranking models commonly employ a "modified pointwise approach," which suffers from theoretical flaws, specifically "objective mismatch" and "noise sensitivity." These issues render models susceptible to the low signal-to-noise ratio inherent in financial markets during optimization. To address these challenges, this study proposes a novel stock ranking prediction framework that integrates a Multi-Layer Perceptron (MLP), Graph Attention Network (GAT), and the Listwise ranking method,LambdaRank.The proposed framework first utilizes an MLP to learn the cross-sectional features of individual stocks. Subsequently, a GAT models the relationships among stocks within the same industry. By adopting LambdaRank as the optimization objective, the framework fundamentally aligns the model's learning direction with the intrinsic nature of the ranking task. Furthermore, this research designs and validates an innovative "top-rank optimization strategy" to guide the model in focusing on improving the prediction accuracy of the most investment-worthy, top-tier stocks. Through rolling-window backtesting on the Taiwan stock market from 2020 to 2024, the proposed model demonstrates exceptional performance, significantly outperforming market benchmarks and baseline model. Ablation studies confirm that both the "top-rank optimization" strategy and the "GNN module" are critical components for the success, with the removal of either leading to a substantial decline in performance. The study also reveals that integrating with models possessing stronger generalization capabilities can effectively mitigate the potential risk of "local overfitting" in this framework when dealing with a larger selection of stocks, thereby constructing a more robust investment decision system with enhanced performance. Overall, this research not only provides a theoretically sound and empirically effective intelligent stock selection framework but also offers profound insights into the design, evaluation, and application of deep learning models in financial markets.
參考文獻 1. Alsulmi, M. (2022). From Ranking Search Results to Managing Investment Portfolios: Exploring Rank-Based Approaches for Portfolio Stock Selection. Electronics (Basel), 11(23), Article 4019. https://doi.org/10.3390/electronics11234019 2. Burges, C. J. C. (1998). A tutorial on support vector machines for pattern recognition. Data Mining and Knowledge Discovery, 2(2), 121–167. https://doi.org/10.1023/A:1009715923555 3. Burges, C. J. C. (2010). From RankNet to LambdaRank to LambdaMART: An overview (Technical Report MSR-TR-2010-82). Retrieved from Microsoft Research website: https://www.microsoft.com/en-us/research/wp-content/uploads/2016/02/MSR-TR-2010-82.pdf 4. Cheng, H.-T., Koc, L., Harmsen, J., Shaked, T., Chandra, T., Aradhye, H., Anderson, G., Corrado, G., Chai, W., Ispir, M., Anil, R., Haque, Z., Hong, L., Jain, V., Liu, X., & Shah, H. (2016). Wide & Deep Learning for Recommender Systems. https://doi.org/10.48550/arxiv.1606.07792 5. Chen, T., & Guestrin, C. (2016). XGBoost: A Scalable Tree Boosting System. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 13-17-, 785–794. https://doi.org/10.1145/2939672.2939785 6. Dash, R. K., Nguyen, T. N., Cengiz, K., & Sharma, A. (2023). Fine-tuned support vector regression model for stock predictions. Neural Computing & Applications, 35(32), 23295–23309. https://doi.org/10.1007/s00521-021-05842-w 7. Fey, M., & Lenssen, J. E. (2019). Fast Graph Representation Learning with PyTorch Geometric. https://doi.org/10.48550/arxiv.1903.02428 8. Feng, F., He, X., Wang, X., Luo, C., Liu, Y., & Chua, T.-S. (2019). Temporal Relational Ranking for Stock Prediction. ACM Transactions on Information Systems, 37(2), Article 27. https://doi.org/10.1145/3309547 9. Gastinger, J., Nicolas, S., Stepić, D., Schmidt, M., & Schülke, A. (2021). A study on ensemble learning for time series forecasting and the need for meta-learning. In 2021 International Joint Conference on Neural Networks (IJCNN) (pp. 1-8). Shenzhen, China: IEEE. doi:10.1109/IJCNN52387.2021.9533378 10. Hsu, Y.-L., Tsai, Y.-C., & Li, C.-T. (2022). FinGAT: Financial Graph Attention Networks for Recommending Top-K Profitable Stocks. IEEE Transactions on Knowledge and Data Engineering, 35(1), 1–1. https://doi.org/10.1109/TKDE.2021.3079496 11. Han, Y., Kim, J., & Enke, D. (2023). A machine learning trading system for the stock market based on N-period Min-Max labeling using XGBoost. Expert Systems with Applications, 211, Article 118581. https://doi.org/10.1016/j.eswa.2022.118581 12. Kumbure, M. M., Lohrmann, C., Luukka, P., & Porras, J. (2022). Machine learning techniques and data for stock market forecasting: A literature review. Expert Systems with Applications, 197, Article 116659. https://doi.org/10.1016/j.eswa.2022.116659 13. Lundberg, S. M., & Lee, S.-I. (2017). A unified approach to interpreting model predictions. In Proceedings of the 31st International Conference on Neural Information Processing Systems (pp. 4768–4777). Curran Associates Inc. https://proceedings.neurips.cc/paper/2017/file/8a20a8621978632d76c43dfd28b67767-Paper.pdf 14. López de Prado, M. (2018). Advances in financial machine learning. Hoboken, NJ: John Wiley & Sons. 15. Ma, T., & Tan, Y. (2022). Stock Ranking with Multi-Task Learning. Expert Systems with Applications, 199, Article 116886. https://doi.org/10.1016/j.eswa.2022.116886 16. Narendra Babu, C., & Eswara Reddy, B. (2015). Prediction of selected Indian stock using a partitioning–interpolation based ARIMA–GARCH model. Applied Computing & Informatics, 11(2), 130–143. https://doi.org/10.1016/j.aci.2014.09.002 17. Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., Desmaison, A., Köpf, A., Yang, E., DeVito, Z., Raison, M., Tejani, A., Chilamkurthy, S., Steiner, B., Fang, L., … Chintala, S. (2019). PyTorch: An Imperative Style, High-Performance Deep Learning Library. https://doi.org/10.48550/arxiv.1912.01703 18. Patel, H., & Sahni, S. (2022). Exploring Explainability Methods for Graph Neural Networks. https://doi.org/10.48550/arxiv.2211.01770 19. Qiao, Y., Xia, Y., Li, X., Li, Z., & Ge, Y. (2023). Higher-order Graph Attention Network for Stock Selection with Joint Analysis. https://doi.org/10.48550/arxiv.2306.15526 20. Rahangdale, A., & Raut, S. (2019). Machine Learning Methods for Ranking. International Journal of Software Engineering and Knowledge Engineering, 29(6), 729–761. https://doi.org/10.1142/S021819401930001X 21. Strader, T. J., Rozycki, J. J., ROOT, T. H., & Huang, Y.-H. J. (2020). Machine Learning Stock Market Prediction Studies: Review and Research Directions. Journal of International Technology and Information Management, 28(4), 63–83. https://doi.org/10.58729/1941-6679.1435 22. Song, G., Zhao, T., Wang, S., Wang, H., & Li, X. (2023). Stock ranking prediction using a graph aggregation network based on stock price and stock relationship information. Information Sciences, 643, Article 119236. https://doi.org/10.1016/j.ins.2023.119236 23. Scholkemper, M., Wu, X., Jadbabaie, A., & Schaub, M. T. (2024). Residual Connections and Normalization Can Provably Prevent Oversmoothing in GNNs. https://doi.org/10.48550/arxiv.2406.02997 24. Veličković, P., Cucurull, G., Casanova, A., Romero, A., Lio, P., & Bengio, Y. (2017). Graph attention networks. arXiv preprint arXiv:1710.10903. 25. Wu, Z., Pan, S., Chen, F., Long, G., Zhang, C., & Yu, P. S. (2021). A Comprehensive Survey on Graph Neural Networks. IEEE Transaction on Neural Networks and Learning Systems, 32(1), Article 9046288. https://doi.org/10.1109/TNNLS.2020.2978386 26. Xiang, S., Cheng, D., Shang, C., Zhang, Y., & Liang, Y. (2022). Temporal and Heterogeneous Graph Neural Network for Financial Time Series Prediction. Proceedings of the 31st ACM International Conference on Information & Knowledge Management, 3584–3593. https://doi.org/10.1145/3511808.3557089 27. Zhang, L., Yin, Y., You, Y., Hajiyev, A., Xu, J., Duca, G., García Márquez, F. P., Ali Hassan, M. H., & Altiparmak, F. (2021). Portfolio Models Based on Fundamental Analysis Using Learning to Rank. In Proceedings of the Fifteenth International Conference on Management Science and Engineering Management (Vol. 79, pp. 352–365). Springer International Publishing AG. https://doi.org/10.1007/978-3-030-79206-0_27 28. Zhang, W., Chen, Z., Miao, J., & Liu, X. (2022). Research on Graph Neural Network in Stock Market. Procedia Computer Science, 214(C), 786–792. https://doi.org/10.1016/j.procs.2022.11.242
描述 碩士
國立政治大學
風險管理與保險學系
112358015
資料來源 http://thesis.lib.nccu.edu.tw/record/#G0112358015
資料類型 thesis
dc.contributor.advisor 黃泓智zh_TW
dc.contributor.author (Authors) 鄭玉海zh_TW
dc.contributor.author (Authors) Cheng, Yu-Haien_US
dc.creator (作者) 鄭玉海zh_TW
dc.creator (作者) Cheng, Yu-Haien_US
dc.date (日期) 2025en_US
dc.date.accessioned 4-Aug-2025 14:11:01 (UTC+8)-
dc.date.available 4-Aug-2025 14:11:01 (UTC+8)-
dc.date.issued (上傳時間) 4-Aug-2025 14:11:01 (UTC+8)-
dc.identifier (Other Identifiers) G0112358015en_US
dc.identifier.uri (URI) https://nccur.lib.nccu.edu.tw/handle/140.119/158518-
dc.description (描述) 碩士zh_TW
dc.description (描述) 國立政治大學zh_TW
dc.description (描述) 風險管理與保險學系zh_TW
dc.description (描述) 112358015zh_TW
dc.description.abstract (摘要) 圖神經網路（GNN）已經在多個領域展現出優良的表現，在股票排名預測中展現了捕捉複雜關聯的潛力，但當前主流模型普遍採用的「改良式 Pointwise 方法」存在理論缺陷，即「目標錯配」與「雜訊敏感」，導致模型在優化過程中易受金融市場低信噪比的影響。為解決此問題，本研究提出一個結合多層感知器（MLP）、圖注意力網路（GAT）與 Listwise 排序方法 LambdaRank 的新型股票排名預測框架。該框架首先利用 MLP 學 習個股的橫截面特徵，再透過 GAT 建模同產業股票間的關聯性，並以 LambdaRank 為優化目標，從根本上使模型的學習方向與排序任務的本質保持一致。此外，本研究設計並驗證了一套創新的「頂部排名優化策略」，引導模型聚焦於提升最具投資價值的頂級股票之預測準確性。 透過對台灣股市 2020 年至 2024 年的滾動窗口回測，本研究提出的完整模型展現了卓越的績效，顯著優於市場基準與基線模型。消融實驗進一步證實，「頂部排名優化」策略與「GNN 模塊」均為模型取得成功的關鍵組件，移除任何一者皆會導致績效顯著衰退。研究亦發現，透過與泛化能力更強的模型進行集成，能有效緩解本框架在較大選股數中可能存在的「局部過擬合」風險，從而構建出一個績效更佳且更穩健的投資決策系統。總體而言，本研究不僅提供了一個理論更健全、實證更有效的智能選股框架，也 為深度學習模型在金融市場的設計、評估與應用提供了深刻的洞見。zh_TW
dc.description.abstract (摘要) Graph Neural Networks (GNNs) have demonstrated superior performance across various domains and exhibit significant potential in capturing complex relationships for stock ranking prediction. However, prevailing GNN stock ranking models commonly employ a "modified pointwise approach," which suffers from theoretical flaws, specifically "objective mismatch" and "noise sensitivity." These issues render models susceptible to the low signal-to-noise ratio inherent in financial markets during optimization. To address these challenges, this study proposes a novel stock ranking prediction framework that integrates a Multi-Layer Perceptron (MLP), Graph Attention Network (GAT), and the Listwise ranking method,LambdaRank.The proposed framework first utilizes an MLP to learn the cross-sectional features of individual stocks. Subsequently, a GAT models the relationships among stocks within the same industry. By adopting LambdaRank as the optimization objective, the framework fundamentally aligns the model's learning direction with the intrinsic nature of the ranking task. Furthermore, this research designs and validates an innovative "top-rank optimization strategy" to guide the model in focusing on improving the prediction accuracy of the most investment-worthy, top-tier stocks. Through rolling-window backtesting on the Taiwan stock market from 2020 to 2024, the proposed model demonstrates exceptional performance, significantly outperforming market benchmarks and baseline model. Ablation studies confirm that both the "top-rank optimization" strategy and the "GNN module" are critical components for the success, with the removal of either leading to a substantial decline in performance. The study also reveals that integrating with models possessing stronger generalization capabilities can effectively mitigate the potential risk of "local overfitting" in this framework when dealing with a larger selection of stocks, thereby constructing a more robust investment decision system with enhanced performance. Overall, this research not only provides a theoretically sound and empirically effective intelligent stock selection framework but also offers profound insights into the design, evaluation, and application of deep learning models in financial markets.en_US
dc.description.tableofcontents 第一章、緒論與研究貢獻 1 第二章、研究方法 10 第三章、實證結果 30 第四章、結論與未來研究建議 45 參考文獻 49 附錄 52zh_TW
dc.format.extent 2239742 bytes-
dc.format.mimetype application/pdf-
dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0112358015en_US
dc.subject (關鍵詞) 學習排序zh_TW
dc.subject (關鍵詞) 圖神經網路zh_TW
dc.subject (關鍵詞) 股票排名預測zh_TW
dc.subject (關鍵詞) LambdaRankzh_TW
dc.subject (關鍵詞) 集成學習zh_TW
dc.subject (關鍵詞) 量化投資zh_TW
dc.subject (關鍵詞) Learning to Ranken_US
dc.subject (關鍵詞) Graph Neural Networken_US
dc.subject (關鍵詞) Stock Ranking Predictionen_US
dc.subject (關鍵詞) LambdaRanken_US
dc.subject (關鍵詞) Ensemble Learningen_US
dc.subject (關鍵詞) Quantitative Investmenten_US
dc.title (題名) 圖神經網路於台灣股市長期股票排名預測之應用：基於橫截面數據與優化排序方法zh_TW
dc.title (題名) Graph Neural Networks for Long-Term Stock Ranking Prediction in the Taiwan Stock Market: An Approach Based on Cross-Sectional Data and Optimized Learning to Ranken_US
dc.type (資料類型) thesisen_US
dc.relation.reference (參考文獻) 1. Alsulmi, M. (2022). From Ranking Search Results to Managing Investment Portfolios: Exploring Rank-Based Approaches for Portfolio Stock Selection. Electronics (Basel), 11(23), Article 4019. https://doi.org/10.3390/electronics11234019 2. Burges, C. J. C. (1998). A tutorial on support vector machines for pattern recognition. Data Mining and Knowledge Discovery, 2(2), 121–167. https://doi.org/10.1023/A:1009715923555 3. Burges, C. J. C. (2010). From RankNet to LambdaRank to LambdaMART: An overview (Technical Report MSR-TR-2010-82). Retrieved from Microsoft Research website: https://www.microsoft.com/en-us/research/wp-content/uploads/2016/02/MSR-TR-2010-82.pdf 4. Cheng, H.-T., Koc, L., Harmsen, J., Shaked, T., Chandra, T., Aradhye, H., Anderson, G., Corrado, G., Chai, W., Ispir, M., Anil, R., Haque, Z., Hong, L., Jain, V., Liu, X., & Shah, H. (2016). Wide & Deep Learning for Recommender Systems. https://doi.org/10.48550/arxiv.1606.07792 5. Chen, T., & Guestrin, C. (2016). XGBoost: A Scalable Tree Boosting System. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 13-17-, 785–794. https://doi.org/10.1145/2939672.2939785 6. Dash, R. K., Nguyen, T. N., Cengiz, K., & Sharma, A. (2023). Fine-tuned support vector regression model for stock predictions. Neural Computing & Applications, 35(32), 23295–23309. https://doi.org/10.1007/s00521-021-05842-w 7. Fey, M., & Lenssen, J. E. (2019). Fast Graph Representation Learning with PyTorch Geometric. https://doi.org/10.48550/arxiv.1903.02428 8. Feng, F., He, X., Wang, X., Luo, C., Liu, Y., & Chua, T.-S. (2019). Temporal Relational Ranking for Stock Prediction. ACM Transactions on Information Systems, 37(2), Article 27. https://doi.org/10.1145/3309547 9. Gastinger, J., Nicolas, S., Stepić, D., Schmidt, M., & Schülke, A. (2021). A study on ensemble learning for time series forecasting and the need for meta-learning. In 2021 International Joint Conference on Neural Networks (IJCNN) (pp. 1-8). Shenzhen, China: IEEE. doi:10.1109/IJCNN52387.2021.9533378 10. Hsu, Y.-L., Tsai, Y.-C., & Li, C.-T. (2022). FinGAT: Financial Graph Attention Networks for Recommending Top-K Profitable Stocks. IEEE Transactions on Knowledge and Data Engineering, 35(1), 1–1. https://doi.org/10.1109/TKDE.2021.3079496 11. Han, Y., Kim, J., & Enke, D. (2023). A machine learning trading system for the stock market based on N-period Min-Max labeling using XGBoost. Expert Systems with Applications, 211, Article 118581. https://doi.org/10.1016/j.eswa.2022.118581 12. Kumbure, M. M., Lohrmann, C., Luukka, P., & Porras, J. (2022). Machine learning techniques and data for stock market forecasting: A literature review. Expert Systems with Applications, 197, Article 116659. https://doi.org/10.1016/j.eswa.2022.116659 13. Lundberg, S. M., & Lee, S.-I. (2017). A unified approach to interpreting model predictions. In Proceedings of the 31st International Conference on Neural Information Processing Systems (pp. 4768–4777). Curran Associates Inc. https://proceedings.neurips.cc/paper/2017/file/8a20a8621978632d76c43dfd28b67767-Paper.pdf 14. López de Prado, M. (2018). Advances in financial machine learning. Hoboken, NJ: John Wiley & Sons. 15. Ma, T., & Tan, Y. (2022). Stock Ranking with Multi-Task Learning. Expert Systems with Applications, 199, Article 116886. https://doi.org/10.1016/j.eswa.2022.116886 16. Narendra Babu, C., & Eswara Reddy, B. (2015). Prediction of selected Indian stock using a partitioning–interpolation based ARIMA–GARCH model. Applied Computing & Informatics, 11(2), 130–143. https://doi.org/10.1016/j.aci.2014.09.002 17. Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., Desmaison, A., Köpf, A., Yang, E., DeVito, Z., Raison, M., Tejani, A., Chilamkurthy, S., Steiner, B., Fang, L., … Chintala, S. (2019). PyTorch: An Imperative Style, High-Performance Deep Learning Library. https://doi.org/10.48550/arxiv.1912.01703 18. Patel, H., & Sahni, S. (2022). Exploring Explainability Methods for Graph Neural Networks. https://doi.org/10.48550/arxiv.2211.01770 19. Qiao, Y., Xia, Y., Li, X., Li, Z., & Ge, Y. (2023). Higher-order Graph Attention Network for Stock Selection with Joint Analysis. https://doi.org/10.48550/arxiv.2306.15526 20. Rahangdale, A., & Raut, S. (2019). Machine Learning Methods for Ranking. International Journal of Software Engineering and Knowledge Engineering, 29(6), 729–761. https://doi.org/10.1142/S021819401930001X 21. Strader, T. J., Rozycki, J. J., ROOT, T. H., & Huang, Y.-H. J. (2020). Machine Learning Stock Market Prediction Studies: Review and Research Directions. Journal of International Technology and Information Management, 28(4), 63–83. https://doi.org/10.58729/1941-6679.1435 22. Song, G., Zhao, T., Wang, S., Wang, H., & Li, X. (2023). Stock ranking prediction using a graph aggregation network based on stock price and stock relationship information. Information Sciences, 643, Article 119236. https://doi.org/10.1016/j.ins.2023.119236 23. Scholkemper, M., Wu, X., Jadbabaie, A., & Schaub, M. T. (2024). Residual Connections and Normalization Can Provably Prevent Oversmoothing in GNNs. https://doi.org/10.48550/arxiv.2406.02997 24. Veličković, P., Cucurull, G., Casanova, A., Romero, A., Lio, P., & Bengio, Y. (2017). Graph attention networks. arXiv preprint arXiv:1710.10903. 25. Wu, Z., Pan, S., Chen, F., Long, G., Zhang, C., & Yu, P. S. (2021). A Comprehensive Survey on Graph Neural Networks. IEEE Transaction on Neural Networks and Learning Systems, 32(1), Article 9046288. https://doi.org/10.1109/TNNLS.2020.2978386 26. Xiang, S., Cheng, D., Shang, C., Zhang, Y., & Liang, Y. (2022). Temporal and Heterogeneous Graph Neural Network for Financial Time Series Prediction. Proceedings of the 31st ACM International Conference on Information & Knowledge Management, 3584–3593. https://doi.org/10.1145/3511808.3557089 27. Zhang, L., Yin, Y., You, Y., Hajiyev, A., Xu, J., Duca, G., García Márquez, F. P., Ali Hassan, M. H., & Altiparmak, F. (2021). Portfolio Models Based on Fundamental Analysis Using Learning to Rank. In Proceedings of the Fifteenth International Conference on Management Science and Engineering Management (Vol. 79, pp. 352–365). Springer International Publishing AG. https://doi.org/10.1007/978-3-030-79206-0_27 28. Zhang, W., Chen, Z., Miao, J., & Liu, X. (2022). Research on Graph Neural Network in Stock Market. Procedia Computer Science, 214(C), 786–792. https://doi.org/10.1016/j.procs.2022.11.242zh_TW