透過您的圖書館登入
IP:3.141.21.115
  • 學位論文

以網路探勘協助競爭產品挖掘

A Web Mining Approach for Competing Product Discovery

指導教授 : 楊錦生
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


隨著商業環境中競爭趨勢愈來愈為激烈,各廠商不斷地推出最新的商品,以期能提升在市場上的競爭力。由於新商品推出頻繁,使得市場上商品的數量龐大、更新速度迅速,增加了商品尋找的複雜度。在此背景下,協助消費者或企業有效率且正確地自大量的商品中,找出有興趣的商品及相關的競爭(或替代)商品,成為一項重要的工作。競爭商品挖掘(Competing Product Mining)可以協助消費者在購物前,了解市場上存在的性質相似的競爭商品,並進一步利用意見挖掘(Opinion Mining)的技術來比較這些商品的優劣,協助購買決策的制定。對企業而言,競爭商品挖掘可以協助他們進行環境掃描(Environmental Scanning),即時掌握競爭對手的動態與競爭商品的市場反應。先前研究並沒有為競爭商品挖掘發展出專門的技術,因此,本研究提出利用網路探勘(Web Mining)與文字探勘(Text Mining)之技術,自網際網路中發掘與擷取有用的資源並加以分析,以找出某一特定商品之競爭商品列表。本研究針對提出的技術,實際建置一個競爭商品挖掘網站,以展現所提技術的可行性。此外,本研究也與現今提供商品簡介的網站之競爭商品列表進行比較,以驗證所提系統的效能。

並列摘要


In the global economy era, manufacturers continue to introduce new products in order to enhance their competitiveness. Consequently, there are huge volumes of diverse products compete each other in the market. Effective and efficient monitoring of the competing products of a specific target product is critical and important for manufacturers, consumers, and other stakeholders. From the consumers’ perspective, competing product mining can help them understand the available alternatives of a specific product and accordingly make better purchasing decision. For product manufactures, competing product mining can help the task of environment scanning which concentrates on the immediate grasp of the activities of their competitors so that the appropriate reactions can be conducted. There are some studies focused on the development of automatic systems for mining competing intelligence, specifically competitors of a focal company. However, the problem of competing product mining is not properly addressed yet. As a result, the purpose of this study is to propose an automatic competing product mining technique. Specifically, the Web is adopted as the information source of the proposed competing product mining technique and some well-known Web mining and text mining techniques are incorporated to accomplish the task. We build a prototype system to demonstrate the feasibility of the proposed idea. Moreover, an experiment is designed and conducted to empirically evaluate the performance of the proposed technique.

參考文獻


16.葉泓誠, "利用文字探勘協助公司競爭與合作網路之建構," 元智大學資訊管理學系碩士論文, 2011.
3.Helmut Schmid, "Improvements in Part-of-Speech Tagging with an Application to German,” Proceedings of the ACL SIGDAT-Workshop, 1995.
6.Murthy Ganapathibhotla and Bing Liu, "Mining Opinions in Comparative Sentences," Proceedings of the 22nd International Conference on Computational Linguistics, 2008.
8.Ricardo Baeza Yates and Berthier Ribeiro Neto, Modern Information Retrieval, Addison Wesley, 1999.
11.Zhongming Ma, Gautam Pant and Olivia R.L. Sheng, "Network-based Approach to Mining Competitor Relationships from Online News," Electronic Commerce Research and Applications, 2011, Vol. 10, No. 4, pp.418–427.

延伸閱讀