The Research on Utilization of Heterogeneous Metadata and System Management of Institutional Repositories on Distributed Environment.

近年来，元数据在信息资源共享中的作用越来越重要。然而，由于元数据分布的框架尚未获得，元数据没有被有效地使用。与此同时，机构知识库每天都在增加。但是，系统管理的方法论还没有建立起来。本研究的目的是实验性地建立元数据框架.我们开发了一个元搜索系统，导入了三种类型的元数据。第一个数据是日本图书情报学会的书目数据，称为比布利斯。第二个是筑波大学知识社区研究中心的元数据，称为ULIS-CORE。第三个是通过OAI-PMH（Open Archive Initiative Protocol for Metadata Harvesting）的几个机构的元数据。比布利斯和ULIS-CORE的元素从原始元素映射到OAI-PMH标准元素。我们开发了一个元数据模式设计系统，包括三个功能。第一个功能是提供元数据的对象配置文件。二是支持图式理解。三是模式推荐。该系统可以支持专业人员和非专业人员的技能程度，并可以缩小技能差距。我们为机构知识库开发了一个联合搜索系统。机构知识库收集研究论文或学位论文等知识产品。研究人员需要根据他们的专业访问专门的文件或数据。但是他们不得不一个库一个库地搜索，因为文档分散在多个机构中。该系统为研究人员提供了机构知识库的一站式检索服务，本研究项目澄清了异构元数据处理中的问题，并给出了部分解决方案。

In recent years, role of metadata is of increasing importance for information resource sharing. However, metadata is not used efficiently because framework of metadata distribution is yet to be acquired. Meanwhile, institutional repositories increase from one day to next. But, methodology of the system management of is not established. The purpose of this research is to build metadata framework experimentally.1. We developed a meta-search system that imported three types of metadata. First data is bibliographic data of Japan Society of Library and Information Science, called BIBLIS. Second one is metadata of Research Center for Knowledge Community, University of Tsukuba, called ULIS-CORE. Third one is metadata of several institutions via OAI-PMH (Open Archive Initiative Protocol for Metadata Harvesting). Elements of BIBLIS and ULIS-CORE are mapped from original element to OAI-PMH standard one.2. We developed a metadata schema design system that consists of three functions. First function is to provide object profile of metadata. Second one is to support schema understanding. Third one is to recommend schema. The system can support both professionals and non-professionals in its degree of skills, and can close the skill gap.3. We developed a federated search system for institutional repositories. The institutional repository collects intellectual products such as research papers or dissertations. Researchers need to access specialized documents or data according to their major. But they have to search repository by repository because the documents are dispersed in more than one institution. The system provides one stop search of institutional repositories to researchers.This research project clarified the issues handling heterogeneous metadata, and gave partial solution.