著者

杉木 健二 松原 茂樹 逸村 裕 SUGIKI Kenji MATSUBARA Shigeki ITSUMURA Hiroshi

出版者

名古屋大学附属図書館研究開発室

雑誌

名古屋大学附属図書館研究年報 (ISSN:1348687X)

巻号頁・発行日

vol.5, pp.41-49, 2007-03 (Released:2009-01-05)

---

This paper describes a method for efficiently generating the metadata from the scholarly document on Web. In this study, we utilized websites resources that were accumulated in large quantities by widely collecting scholarly information on Nagoya University. These web resources increasingly become more important as a digital library. For the sake of generating metadata formalized by Dublin Core Metadata Element Set, our method takes the following two steps: First, our method judges the page type of the document by using the proper noun such as the organization name or the researcher name, or the link origin etc. Next, for generating metadata, our method applies the extraction/generation rules on each page type. Dublin Core corresponds to OAI-PMH, which is often used as a protocol for the exchange of the metadata, so it is easy to distribute scholarly information among the organizations by generating the metadata based on this set.