| DC 欄位 |
值 |
語言 |
| DC.contributor | 企業管理學系 | zh_TW |
| DC.creator | 鄭明顯 | zh_TW |
| DC.creator | Ming-Shien Cheng | en_US |
| dc.date.accessioned | 2016-8-16T07:39:07Z | |
| dc.date.available | 2016-8-16T07:39:07Z | |
| dc.date.issued | 2016 | |
| dc.identifier.uri | http://ir.lib.ncu.edu.tw:88/thesis/view_etd.asp?URN=984401019 | |
| dc.contributor.department | 企業管理學系 | zh_TW |
| DC.description | 國立中央大學 | zh_TW |
| DC.description | National Central University | en_US |
| dc.description.abstract | 知識的發展許多時候是從單一領域的技術進化成為另一個領域的應用，並造成趨勢。也就是說跨領域的研究有這樣的現象---從一個領域的熱門技術橫跨融合到另一個領域的應用。因此提供一個可以追蹤跨領域科技的應用趨勢的工具，對所有公司的技術管理者而言是一個很重要的工具。但搜尋相關文獻，卻不見相關研究討論這個現象，研究跨領域的文獻很少探討技術發展趨勢；研究技術發展趨勢的文獻例如專利分析，則少有討論跨領域的應用。
為了達到這目標，本研究採用跨文件集混合模式---此模式可以辨識在文件集中不同的主題:有三種主題，分別稱為背景主題、共同主題與特定主題，另外特別針對中文專利文件做處理，接著採用2~3-gram方法從長的中文字中斷出流行科技的字詞，然後追蹤此流行科技字詞在另一個領域的應用趨勢。為了驗證本研究的方法，特別作了以下實驗:蒐集台灣專利文件的四個類別--- IPC 編碼分別為: H04N, H04B, H04L, 與 G06Q，主要是傳播通訊各領域與資料處理系統的應用等領域。實驗的結果驗證了本研究提出的方法確實可以追蹤跨領域的技術應用趨勢，並且找到了4個熱門的技術應用發展。進一步而言，本方法也可以偵測跨領域的新興科技與深具潛力的未來技術發展，這是本研究的獨特貢獻。 | zh_TW |
| dc.description.abstract | The knowledge developed in one discipline may revolutionize the practice of another discipline. Furthermore, there is a growing tendency towards the fusion of existing technologies and the integration of applications that spread cross different technological areas. Therefore, a tool that can help to track application trend of cross discipline technologies should be valuable to technology officers in all enterprises. However, none of the research pays special attention to identify cross discipline technology application trend. To achieve the goal, the research adopted Cross-Collection Mixture Model (CCMM) originally developed for identifying concepts among collections of documents; three kinds of concepts are retrieved from collections of documents, namely, common, specific and background themes. The proposed method can also work with Chinese patent documents with which the word segmentation systems tend to append core technologies with other characters to form long words. The proposed method applies 2~3-gram to break the long words to find popular terms and applies the popular terms (technologies) to identify cross discipline technology application trend. To verify the effectiveness of the developed method, four categories of Chinese patent documents (IPC classification code: H04N, H04B, H04L, and G06Q) of the WEBPAT Taiwan were collected. The result shows that the proposed method indeed can track across discipline trendy technology and find 4 cross discipline application technologies terms. Furthermore, our method also can detect the emerging technology or technology opportunity which has great potential for new technology development. This is the unique contribution of this study. | en_US |
| DC.subject | 跨領域研究 | zh_TW |
| DC.subject | 追蹤應用趨勢 | zh_TW |
| DC.subject | 專利分析 | zh_TW |
| DC.subject | 跨文件集混合模式 | zh_TW |
| DC.subject | 中文斷詞 | zh_TW |
| DC.subject | Cross-Discipline Research | en_US |
| DC.subject | Tracking Application Trend | en_US |
| DC.subject | Patent Analysis | en_US |
| DC.subject | Cross–Collection Mixture Model | en_US |
| DC.subject | Chinese Word Segmentation | en_US |
| DC.title | 從中文專利文件追蹤跨領域科技應用的趨勢 | zh_TW |
| dc.language.iso | zh-TW | zh-TW |
| DC.title | Tracking Application Trends of Cross-Discipline Technology from Chinese Patent Documents | en_US |
| DC.type | 博碩士論文 | zh_TW |
| DC.type | thesis | en_US |
| DC.publisher | National Central University | en_US |