【作者】 汤其丰；

【导师】 张贵；

【作者基本信息】 中南林业科技大学 ， 森林经理学， 2009， 硕士

【摘要】 元数据对于促进林业信息的管理、共享和使用有着重要的意义,元数据理论在林业信息进一步挖掘中的应用对于林业信息研究发展有着极其深远的研究意义。通过元数据理论来描述林业信息,提供有关林业信息生产单位数据存储、数据分类、数据内容、数据质量、数据格式、数据交换网络及数据销售等方面的信息,对原始数据集提供概要的描述,促进林业信息共享,为我们更好的利用和改造森林资源提供了可能。通过元数据理论来建立林业信息文档和索引信息,帮助林业生产单位有效地管理和维护林业信息,提高工作效率。基于国家科技支撑计划“森林资源监测技术研究”课题(任务书编号：2006BAD23B0204-3),本论文以广州市林业信息为例主要就元数据理论在广州市林业信息研究中的应用进行探讨。本文以元数据理论为基础,在分析现有林业信息管理运行中存在的问题和当前数据用户需求的基础上,对基于元数据的林业信息管理元数据标准和林业信息利用提出了自己的看法,初步实现了数据库和元数据库的协同管理和信息利用,对基于元数据管理的林业信息的数据挖掘做了一些有益的探索,为今后进一步开展数据挖掘工作奠定了基础。本文研究的主要结论如下：(1)本文基于现行的科学元数据标准和林业科学元数据标准,按照“共享元数据标准、领域元数据标准、领域专用元数据标准”的三层模式,确定了关系类型数据集元数据的UML模型,共包括如下几个模块：分别是：标识信息模块、分发信息模块和维护信息模块等。(2)本文以数据集为单位对历史数据进行整理,按照制定的元数据标准对数据集进行描述并建立元数据库,从整体上对广州市林业信息的元数据理论应用作了尝试,提高了广州市林业信息管理的规范化。(3)实现了数据集的访问和元数据的分类检索和浏览,为用户提供关于数据集的较为详细的描述,如数据产生的方法和条件等各方面的详细记录。(4)在初步实现查询和共享的基础上,本文还对林业信息数据的进一步利用做了一定的数据挖掘,以期提高林业信息的利用效率。元数据理论在林业信息研究中的应用必然是一个逐步推进、逐步深化的过程。我国元数据理论在林业信息研究中的应用在国内的研究正处于起步阶段,各种理念与发达国家相比仍然存在一定的差距,以下问题还有待进一步研究：(1)林业信息中空间数据元数据标准制定。(2)林业信息的权属问题。(3)林业信息数据挖掘的进一步探讨,支撑海量林业信息数据挖掘的分布式计算算法亦待进一步研究。更多还原

【Abstract】 Metadata is of great significance for the promotion of forestry management, sharing and using, metadata theory used in mining forestry information has far-reaching significance for forestry research and development. Through the theory of meta-data to describe the forestry information, providing information which relate to forestry production data storage, data classification, data content, data quality, data formats, data exchange network and data sales, and providing a summary description of the original data set, it promotes the forestry information sharing, provides a possible for better use and transformation of our forest resources. Through Metadata theory, it can establish forestry information documents and index information which help the effective management of forestry production and maintenance of forestry information and enhance efficiency.based on national science and technology support plans "studies of forest resources monitor technical" issue (task book ID:2006BAD23B0204-3), this paper takes Guangzhou forestry information as an main example, exploring the theory of meta-data used in Guangzhou forestry information research.This paper is based on the theory of meta-data and based on the analysis for problems which exist in forestry information management operations and the current needs of data users, puts forward views for forestry information managing metadata standards through the metadata-based and the use of forestry information, initially realizes the management and use of the database and the meta-database, also makes some useful explorations for forestry information data mining which is based on the database management and makes the foundation for the future data mining work.The main conclusions of this paper are as follows:(1) Based on existing scientific meta-data standards and metadata standards of forestry, in accordance with the three-tier model of "meta-data-sharing standards, the area of metadata standards, the field of special meta-data standard" to determine the UML model of the data set meta-data, which includes the following modules: identification information module, the information distribution module and the information maintain module.(2)This paper takes a data set to collate historical data, in accordance with the development of metadata standards to describe data sets and build a metadata database, and makes a try for the metadata theory applications of forestry information of Guangzhou and improves the standardization of forestry information management in Guangzhou.(3) This paper realizes the access of a data set and the classification retrieval and browsing of metadata, providing users with a more detailed description on data sets, such as detailed records of the methods and conditions of data generated.(4) On the basis of initially query and sharing, this article makes a mining for the further use of forestry information data to improve the efficiency of the use of forestry information.The use of meta-data theory in forestry information research certainly is a process of step by step, and gradually deepen. The use of meta-data theory in our forestry information research is still at an initial stage, compared with developed countries, the concept is still existing some gap,. The following questions remain to be further study:(1)making out a spatial metadata standards of forestry information. (2) the ownership of the forestry information. (3)further exploration for forestry information data mining and distributed computing algorithm for massive forestry information data mining.更多还原