【作者】 龚蛟腾；

【导师】 陈能华；

【作者基本信息】 湘潭大学 ， 图书馆学， 2005， 硕士

【摘要】 网络信息海量无序剧增,具有不均匀性和潜在的失控性。大量而无序的信息,不是财富而是灾难。搜索引擎检索效率不甚理想,漏检误检率较高;返回信息量太大,非文本文件检索困难;“一个搜索适用所有用户”的模式,不能满足用户个性化信息需求。于是,用元数据描述网络资源,即网络信息编目,日益受到人们的重视。元数据是关于数据的数据,MARC 和DC 都是描述网络信息的元数据格式。通过修订AACR2 与ISBD(ER)、新增856 等字段和用SGML 或XML 重新包装,MARC 已成为精致的网络信息资源编目方式;DC 是简单描述网络资源的首选,15 个核心元素可选择、可重复和可任意排列,还可用语言、模式和类型来限定与细化元素语义。此外,DC 格式具有弹性和扩展性,允许对元素集进行修改,增删著录信息。发达国家网络信息编目的研究与应用已有重大突破,譬如OCLC InterCat 的MARC 记录达4.5 万条,OCLC CORC 采用MARC和DC 描述网络资源并建立示范性数据库InterCat 和NetFirst,OCLC Spectrum 系统可将各种元数据转换为USMARC,MARCit 可从浏览器“抓取”网站信息并在OPAC 上自动建立MARC 记录。然而,我国网络信息编目还处于起步阶段,应用领域基本上是空白,研究领域以介绍西方国家的成果为主。目前,国家图书馆、北京大学图书馆和上海图书馆等积极探讨中文元数据标准的制定。网络信息数量庞大、质量不一,其编目层次等级、格式要求和描述详略等各不相同,这要求以不同的元数据标准对网络信息资源编目。构建基于DC-MARC 的元数据集,既可对网络信息进行不同等级的编目,又可实现元数据的互相转换。更多还原

【Abstract】 Network information, for its increasing intricately and severely, is unbalanced and out of control. A lot of unordered information is not the wealth, but the disaster. The Search engine is not perfect because of the low retrieval efficiency, the high rate of mistakable or missing outcome, the large amount of returned information and the unsolved non-text document retrieval. The mode of “one search for all user”, can’t meet the user’s individualized message requirements. So, the describing network resource in metadata, namely the network information catalogue, is attached importance to by people day by day. Metadata is the data about data. MARC and DC are both metadata forms that describe network information. By revising AACR2 and ISBD(ER), increasing 856 field and repacking with SGML or XML, MARC has already become the exquisite method of network information catalogue. Briefly describing network resources, DC is the first-selection. DC’s 15 elements can be chosen, be repeated and be arranged at will, still can be limited with language, mode and type. In addition, DC form is elastic and expanding, allowing revising the element, adding or deleting the information items. In developed country, there is already a great breakthrough in the study and application of the network information catalogue. For example, OCLC InterCat’s MARC record is up to 45,000, OCLC CORC adopts MARC and DC to describe network resources and sets up the demonstrative database—InterCat and NetFirst, OCLC’s Spectrum system can change data of various metadata into USMARC data, MARCit can “pick”website’s information from the browser and set up MARC record on OPAC automatically. However, the network information catalogue in our country is still at the starting stage, whose application is basically a blank, whose research mainly introduces the western countries’achievements. At present, National library, Beijing University library and Shanghai library probe actively into the formulation of Chinese metadata standard. There is a large quantity of network information with different qualities, whose catalogue grades, form requirements and description details are all different, which requires cataloguing different network information with different metadata standards. The metadata collection based on DC-MARC, can realize network information catalogue in all level grades and the data conversion of different metadata.更多还原