本研究目的是以類神經網路與對數迴歸模型分析影響台灣製造業創新阻礙之因素,資料來源取自第二次台灣產業創新調查資料庫,抽取該資料庫中2042家台灣製造業公司樣本進行研究,且針對台灣製造業在創新過程中所受到阻礙的影響因素有哪些?以及分析這些因素所帶來的影響程度。本研究則利用類神經網路模型中的多層感知模型 (Multi-Layer Perceptron, MLP) 、線性結構模式 (Linear Structural Relations, LISREL) 與對數迴歸模型 (Logarithmic Regression Model) 作為本研究之架構模型與方法,並進一步探討行銷創新、組織創新、創新支出、保護措施、專利申請這五大變數與創新阻礙之成本因素、知識因素、市場因素、其他 (基礎設施與法規) 因素間的影響;而本文的研究結果則顯示,這五大自變數與創新阻礙的四項應變數間皆有顯著相關性。其結果為行銷創新、組織創新、保護措施、申請專利對創新阻礙的影響皆為顯著正向影響。創新支出之相關活動對創新阻礙的影響皆為顯著負向影響。 最後則提出本研究之貢獻與限制,本研究貢獻則是相較過去研究所不同的在於探討阻礙創新之相關因素,藉由這些因素去了解企業會受到哪些影響及阻礙程度的高低;研究限制則是由於較少研究針對此議題,因此在採集文獻資料時較無直接可印證之相關文獻。
This study analyzes the factors impact on innovational obstruction of manufacturing industry in Taiwan by applying to Network analysis and Logarithmic Regression model. The data of this study is from Second Taiwan Industrial Innovation Survey and this study focuses on tow thousand and forty-two firm-level samples in manufacturing industry. This study utilizes the Multi-Layer Perceptron (MLP), Linear Structural Relations (LISREL) and Logarithmic Regression Model to explore marketing innovation, organizational innovation, innovation expenditures, and protection measures as independent variables and cost, knowledge, market, infrastructure and regulations as dependent variables. The results of this study show that five independent variables and four dependent variables have significant correlation relationships. The results for the marketing innovation, organizational innovation, protection measures, patent applications all of significantly positive impact on innovational obstruction. Innovation expenditures related activities all of significantly negative impact on innovational obstruction. Finally, this study provides the contributions and limitations. The result of this research in comparison to other previous tests indicates a whole new level of innovation. Analytical speaking, by looking at the factors on impacting the change on how much influential it cost, this business show both signs of increasing and decreasing. Moreover, the study is limited for a less focus that only occurs on our investigation team; consequently, collected data has no relevant proof indeed.
為了持續優化網站功能與使用者體驗,本網站將Cookies分析技術用於網站營運、分析和個人化服務之目的。
若您繼續瀏覽本網站,即表示您同意本網站使用Cookies。