臺灣博碩士論文加值系統

近年來，隨著網際網路的發達以及討論社群的普及，越來越多投資經驗豐富的投資者與分析師開始會在網路股票討論社群上發表個股相關預測輿論，而這些預測輿論往往吸引大批投資大眾的關注與跟隨；然而，預測準確的個股推薦輿論將會帶給投資大眾好的投資績效，預測不準確的個股推薦輿論則會使得投資者之投資績效不佳，甚至造成血本無歸的窘境。因此，有效地驗證網路股票討論社群中投資者與分析師所推薦的個股預測輿論已成為廣大投資者未來關注與跟隨個股推薦者之重要選擇參考依據。
因此，本研究主要目的在於針對網路股票討論社群中投資者或分析師對個股的未來走勢推薦買進或賣出的預測輿論進行績效驗證並建立該個股之推薦者的個人績效評等。針對上述目的，本研究主要研究項目包括：(i)網路股票討論社群個股推薦輿論之擷取，(ii)個股推薦績效之驗證與評等方法發展，與(iii)個股推薦績效驗證與評等系統實作。
本研究成果包含發展一網路股票討論社群個股推薦輿論之擷取演算法、建置一網路股票討論社群個股推薦輿論之實體關聯模型(E-R Model)、發展一個股推薦績效之驗證與評等方法，並實作一個股推薦績效驗證與評等系統，以提供投資大眾未來關注與跟隨個股推薦者之選擇參考依據，進而提昇個股投資績效與獲利。

In recent years, with the advance of the Internet and the popularity of social media, more and more investors and analysts who have extensive experience begin publishing sotck-related predictions opinion in the stock discussion social media, which often attract the attention and follow-up of a large number of individual investors; however, the recommendation opinion of individual stocks which predicted accurately will give rise to individual investors a good invest performance, the recommendation opinion of individual stocks which predicted erroneously will give rise to individual investors a shady invest performance, or even cause the river of no return. Therefore, effectively validating the stocks predictions opinion recommeneded by investors and analysts in the stock discussion social media has become the reference basis for investors’ future attention and following individual stock recommender.
Therefore, the main purpose of this study is to conduct a performance verification of the investors’ or analysts in the stock discussion social media based on the opinion of the future recommended buy or sell of stocks and establish the personal performance rating of the recommender of the stock.This objective is achieved by the following: (i) retrieval of recommended public opinions for individual stocks in online discussion communities, (ii) development of the method for validating and evaluating the recommended performance of individual stocks, (iii) the results of implementing a prototype of a system for recommended performance validation and evaluation of individual stocks and experiment with an illustrative example.
This research results includes establishment of the algorithm, the Entity-Relationship Model, development of the method for validating and evaluating the recommended performance and implementing the system. Provide investors the future attentions and follow-up to the recommendation of individual stock selection basis, and thus enhance the performance of individual stocks and profits.

中文摘要 I
ABSTRACT II
誌謝 III
目錄 IV
表目錄 VI
圖目錄 VII
壹、緒論 1
一、研究背景與動機 1
二、研究目的 2
三、研究架構 2
貳、文獻回顧 3
一、移動平均線相關文獻 3
二、預測時距相關文獻 3
三、分析師相關文獻 4
參、網路股票討論社群個股推薦輿論之擷取 6
一、網路股票討論社群個股推薦輿論之擷取 6
二、網路股票討論社群個股推薦輿論之資料塑模 8
肆、個股推薦績效之驗證與評等方法發展 12
一、個股推薦績效之驗證 12
二、個股推薦績效之評等 14
伍、個股推薦績效驗證與評等系統實作 16
一、網路股票輿情內容資料蒐集 16
二、實作畫面呈現 30
陸、結論與未來研究 36
參考文獻 37

[1]行政院主計總處，http://www.dgbas.gov.tw/mp.asp?mp=1。
[2]臺灣證券交易所，http://www.twse.com.tw/zh/。
[3]JIEBA斷詞系統，https://github.com/fxsjy/jieba/blob/master/extra_dict/dict.txt.big。
[4]金慶平，周彥妤，李佳玲，2015，“機構投資者持股與分析師跟隨對企業股權價值評估之攸關性”，當代會計，第十六卷，第二期，頁175-209，十一月。
[5]陳一如，2008，“代表性偏誤影響分析師推薦行為之探討”，中華管理評論國際學報，第十一卷，一期，頁1-32，2月。
[6]樓禎祺，何培基，2003，“股價移動平均線之理論與實務—以台灣股市模擬投資操作為例”，育達研究叢刊，第五、六期，頁27-52，11月。
[7]Achim, K., Altuntas, O., Häusser, T. and Kessler, W., 2011, “Extracting investor sentiment from weblog texts: A knowledge-based approach”, 13th IEEE International Conference on Commerce and Enterprise Computing (CEC 2011), Publ Res Ctr Henri Turdor, Luxembourg, LUXEMBOURG, September 05-07, pp. 1-9.
[8]Clement, M. B., 1999, “Analyst forecast accuracy: Do abbility, resources, and portfolio complexity matter? ”, Journal of Accounting and Economics, Vol. 27, No. 3, pp. 285-303, January.
[9]Duan, J. J. and Zeng, J. P., 2013, “Mining opinion and sentiment for stock return prediction based on web-forum messages”, 10th International Conference on Fuzzy Systems and Knowledge Discovery (FSKD), Shenyang, PEOPLES R CHINA, July 23-25, pp. 984-988.
[10]Frankel, R., Kothari, S. P. and Weber J., “Determinants of the informativeness of analyst research.”, Journal of Accounting and Economics, Vol. 41, No. 1-2, pp. 29-54, April.
[11]Granville, J. E., 1976, Granville's New Strategy of Daily Stock Market Timing for Maximum Profit, Prentice-Hall, Inc.
[12]Hoffer, J. A., Prescott, M. and McFadden, F., Modern Database Management, 8/e, 2006.
[13]Kim, S. H. and Kim D. C., 2014, “Investor sentiment from internet message postings and the predictability of stock returns”, Journal of Economic Behavior & Organization, Vol. 107, No. B, pp. 708-729, April.
[14]Lang, M. H., Lins, K. V. and Miller, D. P., 2004, “Concentrated control, analyst following, and valuation: Do analyst matter most when investors are protected least? ”, Journal of Accounting Research, Vol. 42, No. 3, pp. 589-623, June.
[15]Lang, M. H. and Lundholm, R. J., 1996, “Corporate disclosure policy and analyst behavior. ”, The Accounting Review, Vol. 71, No. 4, pp. 467-492, October.
[16]Luo, B. H., Zeng, J. P. and Duan J. J., 2016, “Emotion space model for classifying opinions in stock message board”, Expert Systems with Applications, Vol. 44, pp. 138-146, February.
[17]O’Brien, P. C., 1988, “Analysts’ forecast as earnings expectations. ”, Journal of Accounting and Economics, Vol. 10, No. 1, pp. 53-83, June.
[18]Sabherwal S., Sarkar S. K. and Zhang Y., 2011, “Do Internet stock message boards influence trading? Evidence from heavily discussed stocks with no fundamental news”, Journal of Business Finance & Accounting, Vol. 38, No. 9–10, pp. 1209-1237, December.
[19]Si J., Mukherjee A., Liu B., Li Q., Li H. and Deng X., 2013, “Exploiting topic based Twitter sentiment for stock prediction”, The 51st Annual Meeting of the Association for Computational Linguistics, Sofia, Bulgaria, August 04-09, pp. 24-29.