What is a dataset?
Metadata Capture is designed around the concepts of metadata, catalogued as datasets, and data lifecycle management. These concepts shape your interactions within Metadata Capture.
What is a dataset?
To understand what a dataset is in Metadata Capture, you must first understand metadata.
Metadata is information that describes data. It provides context and meaning to raw data, making it discoverable and usable. For example:
- A patient record in a hospital system may have metadata describing the record date and data sensitivity level.
- A transport schedule dataset may have metadata describing the route coverage and vehicle types.
In the context of Metadata Capture, a dataset is a catalogued record of this metadata, structured according to the DCAT-AP-LU standards. Key properties of a dataset include the publisher, keywords and access rights. More importantly, the dataset contains metadata about the data, not the data itself.
Think of metadata as a set of details that describes your data, while a dataset is a record of this metadata, structured in compliance with DCAT-AP-LU.
The role of datasets in data sharing
Organisations in all domains—such as in health, environment, and government—need access to accurate and comprehensive data for research, policy making, decision making, and other purposes. For example, a dataset on Luxembourg Population may include age, nationality, and other information needed for social services planning. A dataset on Covid-19 Cases may include vaccination rates, hospitalisations, and other statistics needed for public health responses.
To share and reuse these data effectively, data holders must provide clear and compliant datasets (metadata records) that accurately describe the data they hold.
Data holders and data consumers
Data holders (or data providers) are organisations or entities that collect actual data directly from sources. For example, a hospital may collect health information directly from a patient, or the Ministry of Finance may collect economic data from businesses. In these examples, both the hospital and the Ministry of Finance are data holders.
Data consumers (or data users) are organisations or individuals that need access to data for analysis, research, or decision-making. For example, a researcher may need health data for epidemiological studies, or a government agency may need economic data for policy development. In these examples, both the researcher and the government agency are data consumers.
As a data holder, you are responsible for securely maintaining and disseminating data within your domain. As a data consumer, you are responsible for using data in compliance with applicable laws and regulations.
The role of Metadata Capture
Metadata Capture acts as a bridge between data holders and data consumers. It enables data holders to create and publish well-structured datasets or metadata records—without accessing the underlying data—ensuring information is documented, discoverable, and shared with data consumers in line with data governance and compliance. Read more about Metadata Capture.
Managing datasets
Each dataset in Metadata Capture is unique to its subject and managed independently. Metadata Capture enables you to update, version, or retire datasets as your data changes, supporting the full dataset lifecycle.