2008-04-012024-05-18https://scholars.lib.ntu.edu.tw/handle/123456789/704479摘要：「數位典藏國家型科技計畫」自民國91年起執行，將國家珍貴的典藏，包括文物、標本、文獻資料等加以數位化，迄今已累積豐碩成果，整體典藏可分為16類主題，總計約建置超過200萬筆文物典藏資料。為了讓國外人士也能使用這些豐富的資料，「典藏目錄與內容多語化計畫」擬以五年時間，逐步將數位典藏聯合目錄及典藏計畫成果，以多語化的方式呈現，提供多語介面與內容，以擴大使用層面，提升臺灣典藏和學術資料於國際社會之能見度。 進行典藏目錄與內容多語化的過程中，需考量目錄架構與內容翻譯的問題。過去著重於物件（item-level）的描述，在目錄的呈現上，固然可以詳細查詢各物件資料，但如能對collection level進行描述，可讓使用者迅速明瞭各個collection 的內涵，獲得較為完整的概念。另一方面，資料必須翻譯成多國語言，而能提供外語使用者加以利用；除了以人工方式進行大量翻譯外，如何藉助資訊技術和現有的多語資源，亦為可深入探討之議題。 97年度將遵循前一年所制定的工作流程及規範，除了持續建置collection level description之外，計畫之重心將置於item level metadata欄位（schema）與標籤（tag）的翻譯。利用「數位典藏中外文詞彙」持續進行數位典藏成果資源的清點與翻譯，並與子計畫二密切合作，將翻譯成果匯入數位臺灣文化入口網，與子計畫二共同研討以確定入口網站的知識架構與學科分類之從屬關係。並且在計畫執行的過程中，建立相關的工作流程，以維持計畫在五年間的執行進度與品質。 <br> Abstract: National Digital Archives Program (NDAP) has been initiated since 2002 to promote and coordinate content digitization and preservation at leading museums, libraries, universities, research institutes, and other content holders in Taiwan. During the first phase of the program (2002-2006), more than 2 million records which belong to 16 categories have been established. To broaden public access and promote Taiwan’s heritage to the world, NDAP have set a goal to provide multilingual service as a basis of international promotion and cooperation. Taiwan Digital Archives Multilingual Development Project aims to create the multilingual union catalogue and multilingual content for NDAP in next 5 years. In most cases, the way of resource description is focused on item level metadata. It provides details of an individual object but loses the whole picture of a category, while collection level description provides information about the collection as a whole rather than about individual items within a collection. Collection level description enables content holders to manage collections and to help users selecting collections and searching on the basis of summary description. Translation is absolutely essential on the development of multiliguality. In addition to manual translation, machine translation is another way to try. Cross-language information retrieval enhances the efficiency of public access to heritage holdings. Therefore, developing the multilinguality mechanism and adopting related technologies are the main research issues of this project. The objectives for this subproject in 2008 are：1) continuously working on collection level description and item level metadata schema, 2) translating selected digital archives into difference language, 3) developing the model of multilingual catalogue, 4) building SOP and rules to guarantee quality and progress of the catalogue, collection level description and item level metadata.詮釋資料多語化跨語言資訊檢索數位典藏目錄metadatamultilingualitycross-language information retrievaldigital archives union catalogue數位典藏與數位學習國家型科技計畫‧數位典藏與學習之海外推展暨國際合作計畫-子計畫一：典藏目錄與內容多語化計畫（Ⅱ）