Studying the Impact of Academic ResearchCommercialization on Intelligent LearningThrought Artificial Intelligence Perspective

[[abstract]]第三次人工智慧 (Artificial intelligence, 革命性地崛起，人類科技躍升 智 慧世代 儼然成為進行式。海量數據主要由 智慧型行動個人裝置 經由聯網活動，以及物聯網感知裝置所接收之訊息組成 透過網際網路雙向傳遞，產生各式 結構化、半結構化 與非結構化數據 ；經由人工智慧系統框架演算法，分析海量數據中所蘊藏富含價值之資訊，將其辨識、擷取、分群、歸納，藉此達到預判以及系統自我學習目的，其結果具有高度精準性以及高可信度。各國相繼提出高科技生 產策略，其建構於人工智慧主體的系統 關鍵技術 。 實證 研究指出， 物聯網 (Internet of things, 技術作為鏈結虛擬與實體媒介，能有效雙向傳遞接收之數據以及下達決策指令，且具有可靠性、完整性與即時性。導入產業媒合相容 特性之商業模式， 技術框架具 高信度決策輔助， 能實質 提升產業營運績效。近年 來 科技革新產業面臨轉型或 升級 ，系統框架結合管理學理論模型建構 ，以 學術研究商業化模式 作為 技術 導入 切點，勢必 影響 各行 百業 之 商業模式與經營策略 。 學術研究機構具有前端創新性研究能量，整合 實務 技術 以實現 科技 創新、管理創新 以及價值創新三面向 ，並以擴散創新為主要表現手段 。本研究採以雙架構 量化研究 方式 利用層級分析 Analytic h ierarchy p rocess ,AHP 方法， 建構智慧化層級架構評估準則之優先發展權重排序，並 尋求替代方案之可行性 利用 結構方程模式 Structural equation modeling, SEM 方法， 檢驗 學術研究商業化構念對於智慧製造、智慧學習之結構關聯性 並 驗證本研究提出假說 。 AHP 問卷共回收 146 份有效樣本， SEM 問卷共回收 2 30 份有效樣本。兩份問卷皆獨立且於不同時間序進 行調查，樣本歸類以產業及學界且平均分布狀況良好 ，調查結果具代表性 。研究結 果顯示，層級架構替代方案以智慧學習權重最高，其優先發展 排序第一位，其次為學術研究商業化； 學術研究商業化對智慧製造有正顯著影響，學術研究商業化對智慧學習具正顯著影響，且服務創新於物聯網對學術研究商業化具完全中介影響效果 ，而智慧資本於人工智慧對學術研究商業化不具中介影響效果 。由此可見，研究成果透過學術研究商業化實質影響智慧化應用端。[[abstract]]As the third artificial intelligence ( AI ) revolutionary rises, science and technology suddenly become a generation of wisdom. Massive data is mainly composed of smart mobile personal devices via networking activities and information received by IoT sensor devices, which are transmitted bidirectionally through the Internet to generate a variety of structured, semi structured, and unstructured data; through artificial intelligence systems, the framework algorithm analyzes the valuable information contained in massive data, and identifies, extracts, groups, and summarizes it to achieve the purpose of prejudgment and systematic self learning. The result has high accuracy and high credibility.Various countries have successively proposed high tech production strategies, which are builting on the system's key technologies of artificial intelligence subjects. Empirical research indicates that the Internet of Things (IoT) technology, as a link between virtual and physical media, can transmit and receive data in mutual wayeffectively, and give instructions. It has reliability, integrity and immediacy features. The introduction of a business model with industry matching compatibility and a technical framework with high reliability decision making assistance can substantially improve industrial operation performance. In recent years, the technological innovation industry is facing transformation or upgrading. The system framework is combined with the management theoretical model construction. Theacademic research and commercialization model is used as the technology introduction point, which will inevitably affect the business models and business strategies of various industries. Academic research institutions have front end innovative research energy, integrating practical technology to realize three aspects oftechnological innovation, management innovation and value innovation, and take diffusion innovation as the main forms o f representation.This study adopts a dual-architecture quantitative research methods, and using the analytical hierarchy process (AHP) method to construct a priority development weight ranking of smart hierarchical architecture evaluation criteria, and seeking the feasibility of alternative solutions; using the structural equation model (SEM) method to test the structural relevance of academic research commercialization constructs to intelligence manufacturing and intelligence learning, and verify the hypothesis proposed in this research. There are a total of 146 valid AHP samples and 230 valid SEM samples were collected. Both questionnaires were surveyed independently with different time series. The samples were categorized by industry and academia, and were evenly distributed. The results shows that the survey possesses representativeness.The results of the research shows that the hierarchical architecture alternative has the highest weight in intelligence learning, and it ranked first priority, followed by commercialization of academic research. Commercialization of academic research has a positive impact on intelligence manufacturing, and commercialization of academicresearch has a positive significant relationship with intelligence learning. It also has a significant impact, and service innovation in the Internet of Things has a full mediating effect on academic research commercialization, while intellectual capital inartificial intelligence has no mediating effect on academic research commercialization. It can be seen that the research results influence the intelligent application end through the commercialization of academic research