基於分層注意力網路建構企業破產預測模型

以作者查詢圖書館館藏

、以作者查詢臺灣博碩士

、以作者查詢全國書目

、勘誤回報

、線上人數：19

、訪客IP：18.216.10.195

姓名

劉珊玟(Shan-Wen Liu) 查詢紙本館藏

畢業系所

工業管理研究所

論文名稱

基於分層注意力網路建構企業破產預測模型
(Constructing a Corporate Bankruptcy Prediction Model based on Hierarchical Attention Network)

相關論文

★ 應用失效模式效應分析於產品研發時程之改善	★ 服務品質因子與客戶滿意度關係研究-以汽車保修廠服務為例
★ 家庭購車決策與行銷策略之研究	★ 計程車車隊派遣作業之研究
★ 電業服務品質與服務失誤之探討-以台電桃園區營業處為例	★ 應用資料探勘探討筆記型電腦異常零件-以A公司為例
★ 車用配件開發及車主購買意願探討(以C公司汽車配件業務為實例)	★ 應用田口式實驗法於先進高強度鋼板阻抗熔接條件最佳化研究
★ 以層級分析法探討評選第三方物流服務要素之研究-以日系在台廠商為例	★ 變動良率下的最佳化批量研究
★ 供應商庫存管理架構下運用層級分析法探討供應商評選之研究-以某電子代工廠為例	★ 台灣地區快速流通消費產品銷售預測模型分析研究–以聯華食品可樂果為例
★ 競爭優勢與顧客滿意度分析以中華汽車為例	★ 綠色採購導入對電子代工廠的影響-以A公司為例
★ 以德菲法及層級分析法探討軌道運輸業之供應商評選研究–以T公司為例	★ 應用模擬系統改善存貨管理制度與服務水準之研究-以電線電纜製造業為例

檔案

[Endnote RIS 格式]

[Bibtex 格式]

[相關文章]

[文章引用]

[完整記錄]

[館藏目錄]

[檢視]

[下載]

本電子論文使用權限為同意立即開放。
已達開放權限電子全文僅授權使用者為學術研究之目的，進行個人非營利性質之檢索、閱讀、列印。
請遵守中華民國著作權法之相關規定，切勿任意重製、散佈、改作、轉貼、播送，以免觸法。

摘要(中)

破產預測研究主要關注企業是否可能在未來的一段時間內經歷破產。企業破產不僅對企業本身造成嚴重的損失，還會對社會經濟造成重大影響，如失業率上升、負債擴大等。儘管近年已有一些研究使用人工神經網路模型提取文本特徵以預測企業破產，但人工神經網路模型俗稱黑盒子，其得出的結果缺乏可解釋性。本研究著重於研究美國上市公司破產預測之模型，我們採用分層注意力網路 (Hierarchical Attention Network , HAN) 從10-K年度報告的管理討論與分析 (Management′s Discussion and Analysis , MD&A) 部分提取文本特徵。分層的網路可以反映文本的結構，在單詞和句子級別加入注意力機制使構建文本表示的過程中，依據重要性來區分內容以提供模型的可解釋性。除了使用分層注意力網路建構文本表示，亦運用Bi-LSTM 學習MD&A的年變化量，再結合自S&P Capital IQ 資料庫的公司財務資料做為預測模型的輸入。本研究在實驗中運用貝氏神經網路和人工神經網路對企業破產進行預測，並比較兩種分類模型在不同輸入下的預測性能，但實驗結果並不如預期，推測可能的原因是因為資料噪聲大與模型參數調配不佳。

摘要(英)

Bankruptcy prediction research focuses on whether an enterprise is likely to experience bankruptcy in the future. Enterprise bankruptcy not only causes serious losses to the enterprise itself, but also has a major impact on the social economy, such as rising unemployment and expanding debts. Although there have been some studies in recent years using Artificial Neural Network models to extract text features to predict enterprise bankruptcy, Artificial Neural Network models are commonly known as black boxes, and the results obtained lack interpretability. This study focuses on the model of bankruptcy prediction of listed companies in the United States. We use Hierarchical Attention Network (HAN) to extract text feature from the Management′s Discussion and Analysis (MD&A) part of the 10-K annual report. The layered network can reflect the structure of the text, and the Attention Mechanism is added at the word and sentence levels to make the process of constructing the text representation, distinguishing the content according to the importance to provide the interpretability of the model. In addition to using Hierarchical Attention Networks to construct text representations, Bi-LSTM is also used to learn the annual change in MD&A, and then combined with company financial data from the S&P Capital IQ database as the input of the predicting model. In this study, Bayesian Neural Network and Artificial Neural Network are used to predict the bankruptcy of enterprises in the experiment, and the prediction performance of the two classification models is compared under different inputs, but the experimental results are not as expected. The possible reason is that the large data noise and poor adjustment of model parameters.

關鍵字(中)

★ 破產預測
★ MD&A文本分析
★ 深度學習
★ 分層注意力網路
★ 貝氏神經網路

關鍵字(英)

★ Bankruptcy Prediction
★ MD&A Text Analysis
★ Deep Learning
★ Hierarchical Attention Networks
★ Bayesian Neural Networks

論文目次

摘要 i
Abstract ii
目錄 iii
圖目錄 v
表目錄 vi
第一章、緒論 1
1-1研究背景與動機 1
1-2問題定義 2
1-3研究目的 3
1-4研究方法 3
1-5研究架構 4
第二章、文獻探討 5
2-1破產預測相關研究 5
2-2文本分析 5
2-3分層注意力網路(Hierarchical Attention Network, HAN) 6
2-3-1詞嵌入 (Word Embedding) 8
2-3-2雙向長短期記憶神經網路(Bi-Long Short-Term Memory Neural Network, Bi-LSTM) 9
2-3-3注意力機制 10
2-4貝氏神經網路 (Bayesian Neural Network, BNN) 11
2-4-1貝氏神經網路之不確定性 12
第三章、研究方法 13
3-1原始資料 14
3-2資料前處理 14
3-2-1數值資料 14
3-2-2文本資料 15
3-2-3破產指標 16
3-3模型設計 17
3-3-1詞嵌入 (Module A) 18
3-3-2單詞注意力和句子表示 (Module B) 18
3-3-3句子注意力和文本表示 (Module C) 19
3-3-4 MD&A文本時間變化量 (Module D) 21
3-3-5數值變量 (Module E) 21
3-3-6貝氏神經網路 (Bayesian Neural Network, BNN) 22
3-4模型評估 23
第四章、實驗結果 24
4-1實驗設計 24
4-1-1資料集分割 24
4-1-2輸入變量組合與訓練參數設計 24
4-1-3最佳化模型配置 26
4-2實驗分析與評估 28
4-2-1探討不同輸入組合之預測績效 28
4-2-2探討各輸入組合在不同神經網路下之預測績效 30
4-2-3貝氏神經網路之不確定性 33
4-3文本注意力解釋 34
第五章、結論與未來研究方向 36
參考文獻 38

參考文獻

[1] Almeida, F., & Xexéo, G. (2019). Word embeddings: A survey. arXiv preprint arXiv:1901.09069.
[2] Altman, E. I. (1968). Financial ratios, discriminant analysis and the prediction of corporate bankruptcy. The Journal of finance, 23(4), 589-609.
[3] Arno, H., Mulier, K., Baeck, J., & Demeester, T. (2022). Next-year bankruptcy prediction from textual data: Benchmark and baselines. arXiv preprint arXiv:2208.11334.
[4] Bahdanau, D., Cho, K., & Bengio, Y. (2014). Neural machine translation by jointly learning to align and translate. arXiv preprint arXiv:1409.0473.
[5] Baroni, M., Dinu, G., & Kruszewski, G. (2014). Don’t count, predict! a systematic comparison of context-counting vs. context-predicting semantic vectors. Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 238-247.
[6] Beaver, W. H., Correia, M., & McNichols, M. F. (2012). Do differences in financial reporting attributes impair the predictive ability of financial ratios for bankruptcy? Review of Accounting Studies, 17, 969-1010.
[7] Blundell, C., Cornebise, J., Kavukcuoglu, K., & Wierstra, D. (2015). Weight uncertainty in neural network. International conference on machine learning, 1613-1622.
[8] Cecchini, M., Aytug, H., Koehler, G. J., & Pathak, P. (2010). Making words work: Using financial text as a predictor of financial events. Decision support systems, 50(1), 164-175.
[9] Demoulin, N. T., & Coussement, K. (2020). Acceptance of text-mining systems: The signaling role of information quality. Information & Management, 57(1), 103120.
[10] Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018). Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805.
[11] Graves, A., & Schmidhuber, J. (2005). Framewise phoneme classification with bidirectional LSTM and other neural network architectures. Neural networks, 18(5-6), 602-610.
[12] Gulcehre, C., Chandar, S., & Bengio, Y. (2017). Memory augmented neural networks with wormhole connections. arXiv preprint arXiv:1701.08718.
[13] Holzinger, A., Biemann, C., Pattichis, C. S., & Kell, D. B. (2017). What do we need to build explainable AI systems for the medical domain? arXiv preprint arXiv:1712.09923.
[14] Huang, A. H., Wang, H., & Yang, Y. (2023). FinBERT: A large language model for extracting information from financial text. Contemporary Accounting Research, 40(2), 806-841.
[15] Jang, B., Kim, M., Harerimana, G., Kang, S.-u., & Kim, J. W. (2020). Bi-LSTM model to increase accuracy in text classification: Combining Word2vec CNN and attention mechanism. Applied Sciences, 10(17), 5841.
[16] Janiesch, C., Zschech, P., & Heinrich, K. (2021). Machine learning and deep learning. Electronic Markets, 31(3), 685-695.
[17] Kim, A., & Yoon, S. (2021). Corporate bankruptcy prediction with domain-adapted BERT. EMNLP 2021, 3rd Workshop on ECONLP.
[18] Kumar, P. R., & Ravi, V. (2007). Bankruptcy prediction in banks and firms via statistical and intelligent techniques–A review. European Journal of Operational Research, 180(1), 1-28.
[19] Kusner, M., Sun, Y., Kolkin, N., & Weinberger, K. (2015). From word embeddings to document distances. International conference on machine learning, 957-966.
[20] Li, C., Zhan, G., & Li, Z. (2018). News text classification based on improved Bi-LSTM-CNN. 2018 9th International Conference on Information Technology in Medicine and Education (ITME), 890-893.
[21] Liu, R., Mai, F., Shan, Z., & Wu, Y. (2020). Predicting shareholder litigation on insider trading from financial text: An interpretable deep learning approach. Information & Management, 57(8), 103387.
[22] Liwicki, M., Graves, A., Fernàndez, S., Bunke, H., & Schmidhuber, J. (2007). A novel approach to on-line handwriting recognition based on bidirectional long short-term memory networks. Proceedings of the 9th International Conference on Document Analysis and Recognition, ICDAR 2007.
[23] Lombardo, G., Pellegrino, M., Adosoglou, G., Cagnoni, S., Pardalos, P. M., & Poggi, A. (2022). Machine Learning for Bankruptcy Prediction in the American Stock Market: Dataset and Benchmarks. Future Internet, 14(8), 244.
[24] Mai, F., Tian, S., Lee, C., & Ma, L. (2019). Deep learning models for bankruptcy prediction using textual disclosures. European Journal of Operational Research, 274(2), 743-758.
[25] Mayew, W. J., Sethuraman, M., & Venkatachalam, M. (2015). MD&A disclosure and the firm′s ability to continue as a going concern. The Accounting Review, 90(4), 1621-1651.
[26] Mikolov, T., Chen, K., Corrado, G., & Dean, J. (2013). Efficient estimation of word representations in vector space. arXiv preprint arXiv:1301.3781.
[27] Peris, A., & Casacuberta, F. (2015). A bidirectional recurrent neural language model for machine translation. Procesamiento del Lenguaje Natural(55), 109-116.
[28] Shridhar, K., Laumann, F., & Liwicki, M. (2019). A comprehensive guide to bayesian convolutional neural network with variational inference. arXiv preprint arXiv:1901.02731.
[29] Shumway, T. (2001). Forecasting bankruptcy more accurately: A simple hazard model. The journal of business, 74(1), 101-124.
[30] Xia, Y., Chen, H., & Zimmermann, R. (2023). A Random Effect Bayesian Neural Network (RE-BNN) for travel mode choice analysis across multiple regions. Travel Behaviour and Society, 30, 118-134.
[31] Yang, Z., Yang, D., Dyer, C., He, X., Smola, A., & Hovy, E. (2016). Hierarchical attention networks for document classification. Proceedings of the 2016 conference of the North American chapter of the association for computational linguistics: human language technologies, 1480-1489.

指導教授

葉英傑(Ying-chieh Yeh)

審核日期

2023-7-18

推文