跨領域之研究涉及在兩個或多個領域之間轉移知識以創建新的應用。以前的跨領域分析研究需要專家辨識技術術語，但缺乏系統式的識別新興技術的功能，相反的，辨識新興技術的方法中通常不包含跨領域的分析技術。本研究在交叉收集混合模型的基礎上，開發了一個算法（DIET：Intensitive Interdisciplinary Emerging Technology），以從中文專利中識別跨領域技術和新興技術。為了驗證DIET開發算法的有效性，在H04B：傳輸，H04L：數位傳輸和H04N：影像傳輸等三個傳輸相關技術類別和一個應用技術類別G06Q：包含使用資料處理系統或方法的行政，管理，商業，營運，監督或預測的專利文字。每對專利類別之間的ε-分離度指數遠低於5％的門檻值，另外採用該方法確定的新興技術平均增長率為95.08％。相比之下，兩種基準方法確定新興技術，平均增長率為9.57％和51.49％。 ;Interdisciplinarity involves transferring knowledge among two or more disciplines to create new applications. Previous interdisciplinary analysis research has required experts to identify technology terminology and lacked functions for systematically identifying emerging technology. By contrast, the methodologies for identifying emerging technology do not involve conducting interdisciplinary analysis. This research developed an algorithm (DIET: iDentifying Interdisciplinary Emerging Technology) on the basis of a cross-collection mixture model to identify interdisciplinary technology and emerging technology from patents written in Chinese. To verify the effectiveness of the DIET developed algorithm, documents in three transmission-related technology categories, namely H04B: transmission, H04L: transmission of digital information, and H04N: image communication, and one application technology category, G06Q: comprising patents for administration, management, commerce, operation, supervision, or prediction by using data processing systems or methods, were collected from WEBPAT Taiwan. The ?-separator index between every pair of categories was substantially lower than the 5% threshold. The emerging technology identified using the proposed method had an average growth rate of 95.08%. By contrast, two benchmark methods identified emerging technology with an average growth rate of 9.57% and 51.49%. Keywords: Interdisciplinary analysis, emerging technology identification, Chinese patent analysis, cross-collection mixture model
