- 學位論文
透過網路資料探勘與專利分析建構品質機能展開 - 以智慧型手機為例
Construction of Quality Function Deployment through Web Data Mining and Patent Analysis- The Case of Smart Phones
摘要
在消費性電子產品的市場中，技術不斷推陳出新，良好的研發策略必須與顧客需求相契合，若企業能夠滿足顧客的期望，將能提升品牌價值與顧客消費忠誠度，在全球化的競爭中保有優勢。然而，顧客的喜好多變且難以捉摸，精準地掌握顧客需求對企業來說至關重要。隨著網際網路的普及，許多社群平台因應而生，顧客能透過網路直接表達他們對產品的想法與意見，為集體且動態的資訊，是企業在擬定研發策略時的重要參考。品質機能展開法考量顧客心聲與產品特性之關聯性，將顧客需求轉換成技術需求，有助於對於研發策略的擬定。本研究在選定案例產品之後，運用網路資料爬蟲軟體，擷取電子商務平台上的顧客評論，接著運用文字探勘技術提取負面評論中之關鍵字，並制定系統化的關鍵字篩選規則，挑選出負面評論中具代表性的關鍵字。接著建構以產品元件為基之本體論，定義出產品的關鍵功能，並蒐集關鍵功能之相關文獻。接著運用潛在語意分析法計算出顧客心聲與關鍵功能相關文獻之關聯性，並計算出顧客心聲與關鍵功能的重要度排序，建構品質機能展開。技術功效矩陣揭露特定領域之專利分布情形，是一種重要的專利分析工具。本研究進一步納入技術功效矩陣維度在品質機能展開之架構中，藉以洞悉案例企業之專利在產品關鍵功能中的布局情形，不僅提供案例企業在產品改善上的建議，同時提供企業在未來的智財佈局上的參考依據。
並列摘要
With the rapid technological innovation of consumer electronic products, the successful research and development strategies must match customers’ demand. If enterprise can satisfy the expectations of customers, the brand value and customer loyalty will be enhanced and maintain the advantage in global competition. However, the customers’ preferences are elusive, the ability to grasp customer needs accurately is essential to enterprise. With the popularity of the Internet, many community platforms are established, customers can express their opinions and comments toward product directly through the Internet. The comments on the Internet provide the collective and real-time information, which are important reference for enterprise to develop R&D strategies. Quality function deployment (QFD) transfers customers’ voice into technical needs which contributes to the development of R&D strategies. This study uses the web data crawler software to retrieve the voice of customers (VoC) of case products on the e-commerce platform, then applies text mining approach to extract the keywords of the negative comments. Systematic key term filtering rules are set up to select the representative key terms in the negative comments. Component- based ontology of selected product is constructed, and the key functions are defined based on the ontology, the literature related to key functions are collected for QFD construction. Latent semantic analysis algorithm is used to calculate the correlation between VoC and key functions. Thus, the importance of VoC and key functions can be ranked. Technology function matrix is a patent analytic approach which reveals the patent distribution of specific domain in sub-functions and sub-technologies. This study further incorporates the technology function matrix dimension as the framework of the extended QFD to identify the case companies’ patent distributions to key functions of its product development. This research not only provides the suggestions of product improvement, but also serve as a reference for developing IP portfolio strategy.