臺灣博碩士論文加值系統

科技發展日新月異，人們生活中的許多事物已被電腦取代，而在資料分析上也漸漸地利用電腦取代人力做分析與處理。自然語言處理主要針對語言文字進行自動化的處理，再進而讓電腦進行文字探勘中的分析及運算，而其中財務文字資訊的分析與相關預測應用亦屬於自然語言處理及文字探勘的範疇。本研究欲將財務年報中的文字資訊進行預測及分析公司下市的狀態，利用數千家美國上市公司從1996年至2013年之年報中的管理階層討論與分析（Management’s Discussion and Analysis，MD&A）章節作為實驗資料集，藉由文字探勘技術及運用支持向量機（Support Vector Machine，SVM）學習預測模型，並進一步將詞彙與公司下市的關係做相關分析。

This study aims to adopt natural language processing and machine learning techniques to predict the future delisting status of companies by utilizing the textual information in financial reports on Form 10-K. We conduct the experiments on thousands of US companies’ annual reports from year 1996 to 2013, and the experimental results shows that the prediction performance in average achieves around 70% in terms of accuracy among all models. Further analysis and discussions on the learned models are also provided in the thesis.

1 緒論. . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2 文獻探討. . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.1 自然語言處理. . . . . . . . . . . . . . . . . . . . . . 3
2.2 財務上的相關分析與研究. . . . . . . . . . . . . . . . . 5
2.2.1 在財務上的數字分析與應用. . . . . . . . . . . . . . 5
2.2.2 在財務上的文字分析與應用. . . . . . . . . . . . . . 7
3 研究方法. . . . . . . . . . . . . . . . . . . . . . . . . .12
3.1 問題定義. . . . . . . . . . . . . . . . . . . . . . . .12
3.2 特徵值計算. . . . . . . . . . . . . . . . . . . . . . .12
3.3 支持向量機. . . . . . . . . . . . . . . . .. . . . . . 13
3.4 研究框架. . . . . . . . . . . . . . . . . . . . . . . .16
4 實驗18
4.1 資料集. . . . . . . . . . . . . . . . . . . . . . . . .18
4.2 資料處理. . . . . . . . . . . . . . . . . . . . . . . .22
4.2.1 公司財務報告與股票價格之對應. . . . . . . . . . . .22
4.2.2 文字前處理及索引. . . . . . . . . . . . . . . . . .23
4.2.3 實驗設定. . . . . . . . . . . . . . . . . . . . . .23
4.2.4 詞彙篩選. . . . . . . . . . . . . . . . . . . . . .24
4.2.5 評估標準. . . . . . . . . . . . . . . . . . . . . .25
4.3 實驗結果. . . . . . . . . . . . . . . . . . . . . . . .26
5 結論. . . . . . . . . . . . . . . . . . . . . . . . . . . .39
參考文獻. . . . . . . . . . . . . . . . . . . . . . . . . .41

[1] K. Alkhatib, H. Najadat, I. Hmeidi, and M. K. A. Shatnawi. Stock price prediction using k-nearest neighbor (knn) algorithm. International Journal of Business, Humanities and Technology, 3(3):32–44, 2013.
[2] G. G. Chowdhury. Natural language processing. Annual Review of Information Science and Technology, 37(1):51–89, 2003.
[3] C. Cortes and V. Vapnik. Support-vector networks. Machine Learning, 20(3):273–297, 1995.
[4] A. Devitt and K. Ahmad. Sentiment polarity identification in financial news: A cohesion-based approach. In Proceedings of the 45th Annual Meeting of the Association of Computational Linguistics, pages 984–991, 2007.
[5] R. Feldman. Techniques and applications for sentiment analysis. Communications of the Association of Computing Machinery, 56(4):82–89, 2013.
[6] Z. Han. Data and text mining of financial markets using news and social media. University of Manchester, 2012.
[7] H. Ince and T. B. Trafalis. Kernel principal component analysis and support vector machines for stock price prediction. Transactions, 3(6):629–637, 2004.
[8] S. Kogan, D. Levin, B. R. Routledge, J. S. Sagi, and N. A. Smith. Predicting risk from financial reports with regression. In Proceedings of Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics, pages 272–280, 2009.
[9] F. Li. Annual report readability, current earnings, and earnings persistence. Journal of Accounting and Economics, 45(2):221–247, 2008.
[10] T. Loughran and B. McDonald. When is a liability not a liability? textual analysis, dictionaries, and 10-ks. The Journal of Finance, 66(1):35–65, 2011.
[11] J. B. Lovins. Development of a stemming algorithm. MIT Information Processing Group, Electronic Systems Laboratory Cambridge, 1968.
[12] P. Malo, A. Sinha, P. Korhonen, J. Wallenius, and P. Takala. Good debt or bad debt: detecting semantic orientations in economic texts. Journal of the Association for Information Science and Technology, 65(4):782–796, 2014.
[13] C.-J. W. Ming-Feng Tsai and P.-C. Chien. Discovering finance keywords via continuous space language models. ACM Transactions on Management Information Systems, 2016.
[14] W. Nuij, V. Milea, F. Hogenboom, F. Frasincar, and U. Kaymak. An automated
framework for incorporating news into stock trading strategies. IEEE
Transactions on Knowledge and Data Engineering, 26(4):823–835, 2014.
[15] M. F. Porter. An algorithm for suffix stripping. Program, 14(3):130–137,
1980.
[16] X. Y. Qiu, P. Srinivasan, and Y. Hu. Supervised learning models to predict
firm performance with annual reports: an empirical study. Journal of the
Association for Information Science and Technology, 65(2):400–413, 2014.
[17] M. D. Rechenthin. Machine-learning classification techniques for the analysis and prediction of high-requency stock direction. University of Iowa, 2014.
[18] P. C. Tetlock, M. SAAR-TSECHANSKY, and S. Macskassy. More than words: quantifying language to measure firms’ fundamentals. The Journal of Finance, 63(3):1437–1467, 2008.
[19] M.-F. Tsai and C.-J. Wang. Risk ranking from financial reports. In Proceedings of European Conference on Information Retrieval, pages 804–807, 2013.
[20] M. W. Uhl. Explaining US consumer behavior with news sentiment. ACM Transactions on Management Information Systems, 2(2):9, 2011.
[21] C.-J. Wang, M.-F. Tsai, T. Liu, and C.-T. Chang. Financial sentiment analysis for risk prediction. In Proceedings of International Joint Conference on
Natural Language Processing, pages 802–808, 2013.
[22] W. J. Wilbur and K. Sirotkin. The automatic identification of stop words. Journal of Information Science, 18(1):45–55, 1992.