投資者經常需要獲取有用信息幫助他們進行投資決策，但往往成本很高。像是，許多投資者花費了大量時間來閱讀網路上有關股票的新聞與討論，然後根據不同的信息來評估下一個投資的動作，這些網路信息中包含的情緒與股票價格波動息息相關。
因此，我們將以網路新聞為投資者提供方便且有效率的方法獲得輿情資訊。提供預測給投資者做決策並幫助投資者簡單的解讀網路新聞隱藏的情緒，也可以讓投資者改變投資策略、增加對於風險的洞察力。
本文開發了一種新穎的方法對網路新聞進行情感分析，此方法是基於輿情分析結合機器學習，使用支持向量機(SVM)和決策樹建模，預測股價漲跌。實驗資料使用2018/08/01 ~ 2021/07/05 近三年期間的Google的財經新聞與Yahoo的股市資料做為實驗平台收集財經新聞與財經數據，而實驗預測的股票為航運類股為長榮海運(2603)、陽明海運(2609)、萬海航運(2615)，是國內主要的貨櫃航運公司。
實驗結果表明股價漲跌之間的趨勢和財經新聞的情緒存在著牢固的相關性，加入新聞輿情指數的機器學習方法相較於單純只使用財經數據做預測的方法，具有更高的分類準確率，另外本研究發現財經新聞隱藏的悲觀情緒對於股價預測的結果，相較於樂觀的新聞更具有影響力。

Investors frequently need to obtain useful information to help them make investment strategies, but it is often costly. For example, many investors spend a lot of time reading news and discussions about stocks on the Internet and then evaluate their next investment actions based on different information. The sentiment contained in this Internet information is closely related to stock price fluctuations. Therefore, online news will provide investors with a convenient and efficient way to obtain public opinion information, provide forecasts for investors to make decisions, and help investors to simply interpret the emotions hidden in online news, also allow investors to change their investment strategies and Increase insight into risks.
This article develops a novel method for sentiment analysis of online news. This method is based on public opinion analysis combined with machine learning, using Support Vector Machines (SVM) and decision tree modeling to predict the rise and fall of stock prices. The experimental data uses the financial news of Google and the stock market data of Yahoo in the past three years from 2018/08/01 to 2021/07/05 as the experimental platform to collect financial news and financial data. The experiment predicted stocks include Evergreen Shipping (2603), Yangming Shipping (2609), and Wanhai Shipping (2615). They are shipping stocks and also the major container shipping companies in Taiwan.
The experimental results show that there is a strong correlation between the trend of stock price rises and falls and the sentiment of financial news. The machine learning method that adds the news public opinion index has a higher classification accuracy than the method that only uses financial data to make predictions. In addition, this study found that the pessimistic emotion hidden in financial news has more influence on the results of stock price forecasts than optimistic news.

謝辭 I
摘要 II
ABSTRACT III
目錄 IV
圖目錄 VI
表目錄 VII
第一章 緒論 1
1.1 研究背景 1
1.2 研究動機與目的 1
1.3 論文架構 2
第二章 文獻回顧 3
2.1 文字探勘簡介 3
2.2 文字探勘結合股價預測之相關研究 4
2.3 情感分析與股價預測之相關研究 4
2.4 傳統股票預測的方法 5
2.5 其他股票預測方法 6
2.6 情感分析與股價預測之相關研究 7
第三章 研究方法 9
3.1 資料蒐集與樣本選擇 9
3.2 研究設計 9
3.2.1 新聞輿情分析 11
3.2.2 計算輿情指數 12
3.2.3 SVM介紹 13
3.2.4 決策樹介紹 13
3.3 系統架構 14
第四章 實驗結果與討論 17
4.1 實驗資料集 17
4.2評估方法 17
4.2.1 準確率(Accuracy) 17
4.2.2 凱利公式 17
4.3 使用SVM分類器對前5、7、10日資料之預測準確率 18
4.4 使用決策樹與SVM分類器預測漲跌準確率 24
4.5 樂觀與悲觀情緒對預測之影響 31
第五章 結論與未來展望 33
參考文獻 35

[ 1 ] Watkins, B.D. (2003). Riding the Wave of Sentiment: An Analysis of Return Consistency as a Predictor of Future Returns. Journal of Behavioral Finance, 4, 191 - 200.
[ 2 ] Lux, T. (2011). Sentiment dynamics and stock returns: the case of the German stock market. Empirical Economics, 41, 663-679.
[ 3 ] Celov, D., & Grigaliuniene, Z. (2010). Economic Forces, Sentiment and Emerging Eastern European Stock Markets.
[ 4 ] Sheu, H., Lu, Y., & Wei, Y. (2010). Causalities Between Sentiment Indicators And Stock Market Returns Under Different Market Scenarios. The International Journal of Business and Finance Research, 4, 159-171.
[ 5 ] Wang, Y., & Iorio, A.D. (2007). The cross section of expected stock returns in the Chinese A-share market. Global Finance Journal, 17, 335-349.
[ 6 ] Baker, M.P., & Wurgler, J. (2003). Investor Sentiment and the Cross-Section of Stock Returns. Econometrics: Single Equation Models eJournal.
[ 7 ] Zhang, Y., & Swanson, P.E. (2010). Are day traders bias free?—evidence from internet stock message boards. Journal of Economics and Finance, 34, 96-112.
[ 8 ] Ghidini, C., Francescomarino, C.D., Rospocher, M., Tonella, P., & Serafini, L. (2012). Semantics-Based Aspect-Oriented Management of Exceptional Flows in Business Processes. IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews), 42, 25-37.
[ 9 ] Abdous, M., & He, W. (2011). Using text mining to uncover students' technology-related problems in live video streaming. Br. J. Educ. Technol., 42, 40-49.
[ 10 ] Miner, G.D., Elder, J.F., Hill, T., Nisbet, R.A., Delen, D., & Fast, A. (2012). Practical Text Mining and Statistical Analysis for Non-structured Text Data Applications.
[ 11 ] Liu, B., Cao, S., & He, W. (2011). Distributed data mining for e-business. Information Technology and Management, 12, 67-79.
[ 12 ] Sullivan, D. (2001). Document Warehousing and Text Mining: Techniques for Improving Business Operations, Marketing, and Sales.
[ 13 ] Wysocki, P.D. (1998). Cheap Talk on the Web: The Determinants of Postings on Stock Message Boards. Capital Markets eJournal.
[ 14 ] Ahmad, K., Oliveira, P.C., Manomaisupat, P., Casey, M.C., Tugba, & Taskaya (2002). Description of Events : An Analysis of Keywords and Indexical Names.
 
[ 15 ] Huberman, G., & Regev, T. (2001). Contagious Speculation and a Cure for Cancer: A Nonevent that Made Stock Prices Soar. Journal of Finance, 56, 387-396.
[ 16 ] Das, S., & Chen, M.Y. (2007). Yahoo! for Amazon: Sentiment Extraction from Small Talk on the Web. Manag. Sci., 53, 1375-1388.
[ 17 ] Antweiler, W., & Frank, M.Z. (2001). Is All That Talk Just Noise? The Information Content of Internet Stock Message Boards. Journal of Finance, 59, 1259-1294.
[ 18 ] Schumaker, R.P., & Chen, H. (2009). Textual analysis of stock market prediction using breaking financial news: The AZFin text system. ACM Trans. Inf. Syst., 27, 12:1-12:19.
[ 19 ] Seybert, N., & Bloomfield, R. (2009). Contagion of Wishful Thinking in Markets. Management Science, 55(5), 738–751.
[ 20 ] Luo, Xueming. (2007). Consumer Negative Voice and Firm-Idiosyncratic Stock Returns. Journal of Marketing. 71. 75-88.
[ 21 ] Atiya, A.F., Talaat, N., & Shaheen, S.I. (1997). An efficient stock market forecasting model using neural networks. Proceedings of International Conference on Neural Networks (ICNN'97), 4, 2112-2115 vol.4.
[ 22 ] Xiao, W., Fei, Q., & Chen, W. (2008). Forecasting Chinese Stock Markets Volatility Based on Neural Network Combining. 2008 Fourth International Conference on Natural Computation, 7, 23-27.
[ 23 ] Yu, L., & Rong, F.S. (2010). Stock Market Forecasting Research Based on Neural Network and Pattern Matching. 2010 International Conference on E-Business and E-Government, 1940-1943.
[ 24 ] Powell, N.A., Foo, S.Y., & Weatherspoon, M.H. (2008). Supervised and Unsupervised Methods for Stock Trend Forecasting. 2008 40th Southeastern Symposium on System Theory (SSST), 203-205.
[ 25 ] Kelly, J.L. (1956). A new interpretation of information rate. IRE Trans. Inf. Theory, 2, 185-189.
[ 26 ] Wang, Z.M., & Wu, C. (2010). Application of Support Vector Regression Method in Stock Market Forecasting. 2010 International Conference on Management and Service Science, 1-4.
[ 27 ] Xiao-si, X., Ting, C., & Ruo-en, R. (2006). Studying on Forecasting the Enterprise Bankruptcy Based on SVM. 2006 International Conference on Management Science and Engineering, 1041-1045.
 
[ 28 ] Yujun, Y., Yimei, Y., & Jianping, L. (2016). Research on financial time series forecasting based on SVM. 2016 13th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP), 346-349.
[ 29 ] Dai, H., Zhang, Y., & Wang, D. (2011). Price Prediction of Stock Index Futures Based on SVM. 2011 Fourth International Conference on Business Intelligence and Financial Engineering, 54-57.
[ 30 ] Yu, H., & Liu, H. (2012). Improved Stock Market Prediction by Combining Support Vector Machine and Empirical Mode Decomposition. 2012 Fifth International Symposium on Computational Intelligence and Design, 1, 531-534.
[ 31 ] Jiang, Y., Wang, H., & Xie, Q. (2009). Classification model of companies' financial performance based on integrated support vector machine. 2009 International Conference on Management Science and Engineering, 1322-1328.
[ 32 ] Nair, B.B., Dharini, N.M., & Mohandas, V.P. (2010). A Stock Market Trend Prediction System Using a Hybrid Decision Tree-Neuro-Fuzzy System. 2010 International Conference on Advances in Recent Technologies in Communication and Computing, 381-385.
[ 33 ] Zheng, Q., & Yanhui, J. (2007). Financial Distress Prediction Based on Decision Tree Models. 2007 IEEE International Conference on Service Operations and Logistics, and Informatics, 1-6.
[ 34 ] Hasan, M.R., Maliha, M., & Arifuzzaman, M. (2019). Sentiment Analysis with NLP on Twitter Data. 2019 International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering (IC4ME2), 1-4.
[ 35 ] Sathya, V., Venkataramanan, A., Tiwari, A., & Ps, D.D. (2019). Ascertaining Public Opinion Through Sentiment Analysis. 2019 3rd International Conference on Computing Methodologies and Communication (ICCMC), 1139-1143.
[ 36 ] Woldemariam, Y.D. (2016). Sentiment analysis in a cross-media analysis framework. 2016 IEEE International Conference on Big Data Analysis (ICBDA), 1-5.
[ 37 ] Shen, W., & Xing, M. (2009). Stock Index Forecast with Back Propagation Neural Network Optimized by Genetic Algorithm. 2009 Second International Conference on Information and Computing Science, 2, 376-379.
[ 38 ] Aasi, B., Imtiaz, S.A., Qadeer, H.A., Singarajah, M., & Kashef, R.F. (2021). Stock Price Prediction Using a Multivariate Multistep LSTM: A Sentiment and Public Engagement Analysis Model. 2021 IEEE International IOT, Electronics and Mechatronics Conference (IEMTRONICS), 1-8.
 
[ 39 ] Ti, A., Du, Z., & Zhang, W. (2019). Analysis on the Volatility of Sustainable Stock Index and Traditional Stock Index Based on GARCH Model. 2019 International Conference on Economic Management and Model Engineering (ICEMME), 47-50.
[ 40 ] Yin, L., Zhang, N., He, L., & Fang, W. (2016). A Study of Relationship between Investor Sentiment and Stock Price Based on Text Mining. 2016 International Conference on Identification, Information and Knowledge in the Internet of Things (IIKI), 536-539.
[ 41 ] Peng, D. (2019). Analysis of Investor Sentiment and Stock Market Volatility Trend Based on Big Data Strategy. 2019 International Conference on Robots & Intelligent System (ICRIS), 269-272.
[ 42 ] Yu, Y., Wang, S., & Zhang, L. (2017). Stock price forecasting based on BP neural network model of network public opinion. 2017 2nd International Conference on Image, Vision and Computing (ICIVC), 1058-1062.