Preskočite na glavni sadržaj
Login
State Geodetic Administration logo National Spatial Data Infrastructure logo

Metadata

Metadata have been used for a long time, since people started making lists of things, inventories, events and other objects or phenomena for the purpose of their systematic records and getting an overview of them. When using spatial data, such as maps in raster or vector format in a GIS environment, spatial databases, or when downloading data through a web-service, we often don't know or are not sure who collected the data and when they did it, what the quality of the data is, what coordinate reference system the data are in, what version the data format is or other information on data that allow being used and interpreted. Such data and their application have a lower value. To avoid these problems, metadata are collected.

Metadata are an integral part of data and together they make single units. Today, the most general definition of metadata is that they are data about data. Metadata need to be documented during the collection of data and they need to be updated when data are changed. In other words, important actions related to data have to be documented through metadata.

Metadata thus became a general term referring to anything of interest. However, describing different types of resources requires different types of metadata and different metadata thematic norms. Inventarisation of everything (knowledge), made possible by metadata, is supported by computers and the amount of information computer systems can store and process, while the development of IT technologies made data sharing an everyday necessity. To be able to use shared data, different groups of users need to know the format of data, their coordinate reference system and other information that can be obtained through metadata catalogues.

There is a clear trend of developing systematic metadata cataloguing (e.g. INSPIRE and national metadata catalogues). To connect different metadata systems, there needs to be some structural link between metadata. INSPIRE was created under a strong influence of ISO metadata standardization, while Croatian NSDI metadata, except national needs, meet the INSPIRE requirements, as well.

In order for spatial data to be included in the NSDI system, they must have metadata. NSDI metadata enables: identification of spatial data sources, their categorization, determination of it position in space, estimation of its time frame, quality, access and use, whom to contact for access to data and other information. NSDI metadata is elaborated in NSDI Metadata Specification v.3.0. Each NSDI metadata has an element that compares the NSDI metadata with ISO and INSPIRE metadata, which enables the sharing and interoperability of the metadata.

Harvesting, as an operation allowing downloading metadata from a metadata catalogue, enables linking between catalogues and linking several catalogues with applications, as internet search engines do. The Dublin Core Metadata Initiative (DCMI), as a respectable international non profit organization involved in metadata standardization for various professional, thematic and business needs, differentiates between fpur levels of metadata interoperability:
 
  • Shared term definitions defines metadata interoperability based on shared natural-language definitions. Within an application environment such as an intranet, library system, or repository, participants standardize the terms they use in the metadata system. Most existing metadata applications currently operate at this level of operability.
  • Formal semantic interoperability defines metadata interoperability based on the shared formal model provided by Resource Description Framework (RDF), which is developed to support linked data that are exposed, shared, and connecting pieces of data, information, and knowledge on the Semantic Web using URI and RDF. Over the past few years, this level of interoperability has evolved significantly through the integration of public and private sector data into data clouds.
  • Description Set syntactic interoperability defines metadata interoperability through applications compatible with the linked data model and, in addition, share an abstract syntax for validatable metadata records, the "description set".
  • Description Set Profile interoperability defines metadata interoperability as the process by which metadata records are exchanged between applications that use metadata and further share an abstract common set od conditions, use the same dictionary, and reflect a shared model od the world. The last two levels of metadata interoperability are in the developmental or scientific domain.

Besides the basic role of metadata to describe data, those in the system of INfrastructure for SPatial Information (INSPIRE) or National Spatial Data Infrastructure (NSDI) also meet a certain functionality, i.e. business model.

ISO defines metadata as data on data, which is a general definition that does not emphasize the business model that metadata seeks to achieve. INSPIRE defines metadata as information describing spatial datasets and spatial data services allowing their finding and use.

This definition sets the goals for the business model that metadata must achieve (i.e. describe spatial data and find and use spatial data). 

ISO metadata packages

Metadata packages and relations between entities

Metadata for geographic data are represented by means of UML packages. Each package contains one or more entities (UML classes) that can be specified or generalised. Entities contain elements (UML class attributes) that identify discrete metadata elements. Entities can be aggregated and repeated as needed. The figure shows the general overview of a package. Metadata are specified by a UML model diagram and data dictionary for each package

The figure illustrates a general view of the package. Metadata is specified by the UML model diagram and data dictionary for each package.

Relationship between metadata packages and metadata entities are shown in Table 6.1. Metadata packages are shown in column 'Package' and aggregated metadata entities contained within a package are shown in column 'Entity'. Each package includes sub-classes.

Relationship between metadata packages and metadata entities

 
Package Entity
Metadata entity set information MD_Metadata
Identification information MD_Identification
Constraint information MD_Constraints
Data quality information DQ_DataQuality
Maintenance information MD_MaintenanceInformation
Spatial representation information MD_SpatialRepresentation
Reference system information MD_ReferenceSystem
Content information MD_ContentInformation
Portrayal catalogue information MD_PortrayalCatalogueReference
Distribution information MD_Distribution
Metadata extension information MD_MetadataExtensionInformation
Application schema information MD_ApplicationSchemaInformation
Extent information EX_Extent
Citation and responsible party information CI_Citation CI_ResponsibleParty

Package abbreviations

Two-letter abbreviations are used to denote the package that contains a class. Those abbreviations precede class names, connected by a "_".

List of Abbreviations:

 
CC Changing Coordinates (ISO 19111)
CI Citation (ISO 19115)
CV   Coverages (ISO 19123)
DQ Data quality (ISO 19115)
DS Dataset (ISO 19115)
EX   Extent (ISO 19115)
FC Feature Catalogue (ISO 19110)
FE Feature (ISO 19109)
FT Feature Topology (ISO 19107)
GF General Feature (ISO 19109)
GM Geometry (ISO 19107)
GR Graph (ISO 19107)
LI Lineage (ISO 19115)
MD Metadata (ISO 19115)
PF Feature Portrayal (ISO 19117)
PS Positioning Services (ISO 19116)
RS Reference System (ISO 19115)
SC Spatial Coordinates (ISO 19111)
SI Spatial Identification (ISO 19112)
SV Services (ISO 19119)
TM Temporal (ISO 19108)
TP Topology (ISO 19107)
TS Simple Topology (ISO 19107)

Metadata application information

The UML class diagram in the image defines the geographic information classes that the metadata refers to.

The picture shows a UML class diagram that defines the classes of geographic information to which the metadata relates.

Metadata UML model

Metadata for describing geographic data is defined using an abstract object UML model. Each diagram of the subclass defines a metadata section (UML package) of related entities, data types and code lists.  

Entities may have mandatory and/or optional elements and associations. In some cases, optional elements may have mandatory elements. 

More information about ISO UML Diagrams:

ISO UML Diagrams Metadata package 
ISO UML Diagram for identification information
ISO UML Diagram Constraint information
ISO UML Diagram for data quality information
ISO UML Diagram for maintenance information
ISO UML Diagram for spatial representation information
ISO UML Diagram for reference system information
ISO UML Diagram for content information
ISO UML Diagram for portrayal catalogue information
ISO UML diagram for distribution information
ISO UML diagram for metadata extension information
ISO UML diagram for application schema information
ISO UML diagram for extent information
ISO UML diagram for citation and responsible party information