利用文字探勘協助公司競爭與合作網路之建構

A Text Mining Approach for Automatic Discovery of Company Competition/Cooperation Network

葉泓誠(Hong-Cheng Ye)

指導教授 ： 楊錦生

元智大學/資訊學院/資訊管理學系/碩士(2011年)

https://doi.org/10.6838/YZU.2011.00267

摘要

隨著商業環境中的競爭趨勢之演變愈來愈激烈的情況下，對企業來說，競爭智慧(Competitive Intelligence)已成了企業本身最重要的商業資訊之一環。畢竟在全球的公司數目相當之龐大，且跨產業之企業的數量也隨之而成長，企業本身的競爭者已經不侷限在相同產業中，而是存在於跨產業之中。雖然挖掘競爭者固然重要，但聯盟是企業另一個發展趨勢，企業的競爭者有可能會有合作夥伴的加入，成為一個聯盟，故競爭者的合作夥伴即可能是企業本身的潛在競爭者，且在未來也許有機會將其吸收為企業本身的合作者(Cooperator)。另一方面，競爭者與合作者之資訊，對於投資者在設計投資組合上，有一定程度上的幫助，對於企業或投資人在利益上均有利。在先前研究中，主要採用公司網站、搜尋引擎或線上新聞之資料，並利用資料探勘之技術協助(Data Mining)發掘公司之間競爭關係，但卻存在著某些限制問題。因此，本研究提出利用文字探勘(Text Mining)之技術來協助公司競爭與合作網站之建置，排除先前研究限制，並挖掘出新的關係(合作關係)，主要可產生出目標公司之競爭者與合作者之列表結果，最後與現今提供公司資源簡介的網站之競爭者資訊結果做正確性及即時性的比較。

關鍵字

競爭智慧 ； 競爭者發掘 ； 合作者發掘 ； 文字探勘 ； 自然語言處理

並列摘要

In the global economy era, the business environment becomes more turbulent and the uncertainty faced by companies increases. To gain sustainable competitiveness in such competitive and uncertain marketplace, companies need to monitor their environments to produce competitive intelligence that helps managers make crucial business decisions. Automatic discovery of competitors and cooperators of a focal company represents a critical and essential task of competitive intelligence. From the company perspective, competitor and cooperator discovery helps a company understand the important players in its target marketplace(s) and formulate its competition strategies accordingly. Competitor and cooperator discovery is also important for other stakeholders. For example, a comprehensive and up-to-date list of competitors and cooperators of a focal company is essential for shareholders or venture capitals to determine the value of a target company. There are several commercial databases (e.g., Yahoo! Finance, Hoover’s, Mergent) which provide the competitor lists of major companies. Moreover, some prior studies also proposed some approach for the automatic discovery of competitors. However, they all encounter some critical limitations. For example, they neglect the discovery of cooperators (e.g., suppliers, customers) and the strengths of the relationships. In response, this study concentrates on proposing a technique which incorporates some state-of-the-art text mining and natural language processing techniques to address the abovementioned limitations and then automatically discover competitors and cooperators of a focal company on the basis of some news collected from the Web. The proposed technique is empirically evaluated using eight well-known multinational companies to demonstrate its effectiveness. The evaluation results are promising and will serve as a basis for future studies.

並列關鍵字

Natural Language Processing ； Competitive Intelligence ； Competitor Mining ； Cooperator Mining ； Text Mining

參考文獻

4. Gary Hamel, Yves L. Doz, C.K. Prahalad, Joel Bleeke, David Ernst, Ashish Nanda, Peter J. Williamson, Benjamin Gomes-Casseres, Rosabeth Moss Kanter, Carlos Ghosn, Henry W. Chesbrough and David J. Teece, (2006). "當對手變夥伴", 天下文化 -財經企管, 蔡慧菁(譯).

6. Zhongming Ma, Olivia R.L. Sheng and Gautam Pant (2009). "Discovering company revenue relations from news: A network approach." Decision Support Systems, Vol.47, No.4, pp.408-414.

11. Sergey Brin and Lawrence Page (1998). "The anatomy of a large-scale hypertextual Web search engine." Computer Networks and ISDN Systems, Vol.30, No.1-7, pp.107-117.

12. JON M. KLEINBERG (1999). "Authoritative sources in a hyperlinked environment." Journal of the ACM, Vol.46, No.5, pp.604-632.

13. Ulrik Brandes (2001). "A Faster Algorithm for Betweenness Centrality." Journal of Mathematical Sociology, Vol.25, No.2, pp.163-177.