UN Semantic Interoperability Framework for Normative and Parliamentary Documents (UNSIF)

The high volume of documents currently produced by the different UN organizations, as well as the amount of non-UN documents that have to be processed, is posing a significant challenge to the UN System. Effective and efficient information management and accessibility is well beyond the “human processing” capabilities of UN organizations.

To address these challenges, UN organizations must overcome the paper paradigm, where documents are “ designed for humans to read, not for computer programs to manipulate meaningfully”, moving to a new form of content “ meaningful to computers [that] will unleash a revolution of new possibilities” (Sir Tim Berners-Lee, 2001).

Between June 2016 and March 2017 the WGDS met online and offline to carry out joint analyses and consultations geared to the definition of a common document standard for the UN system. The work was co-led by the United Nations Department for General Assembly and Conference Management, the Food and Agriculture Organization and the High Level Committee on Management.

In late March 2017, the WGDS recommended to the High Level Committee on Management the adoption of the  UN Semantic Interoperability Framework for normative and parliamentary documents (UNSIF). At the Thirty-Third Session of the High Level Committee on Management (HLCM) held in Budapest, 30-31 March 2017, the HLCM adopted the UN Semantic Interoperability Framework for normative and parliamentary documents (UNSIF)

Why UNSIF?

The adoption of UNSIF has put the basics for the establishment of a UN-wide ecosystem of documents that will foster collaboration and reduce costs in information management across the system by transforming the web of information enclosed in traditional word processing documents into a web of data that can be interpreted by computers to create innovative services.

Having documents available in a common semantically rich format will raise dramatically the UN capacity to coordinate activities, create synergies and respond to clients’ information demands, improve the efficiency and quality of information processing, dissemination and accessibility, and bring significant benefits in terms of governance, accountability and transparency.

UNSIF represents the starting point is for a new era of information management and will lead to considerable efficiencies in document management by delivering productivity gains in the translation processes, promoting multilingualism and allowing for greater accessibility to UN documents by persons with disabilities. Having machine-readable documents available in a common semantically rich format will also be a considerable asset for the implementation of Agenda 2030, which requires a robust review mechanism and a solid framework for evidence-based policies and accountability.

Components

Akoma Ntoso for the United Nations (AKN4UN)

Akoma Ntoso: XML OASIS standard to represent parliamentary, legislative and judiciary documents in a machine-readable format. A format not only ideal for multi-channel dissemination and longtime digital preservation but also to support machine processing to enable the development of smart information services. For the purposes of UNSIF, Akoma Ntoso has been localized to respond the specific requirements of UN parliamentary and normative documents, resulting in Akoma Ntoso for the United Nations System (AKN4UN).

AKN4UN Specifications (Part I)

AKN4UN Document Modelling (Part II)

United Nations Document Ontology (UNDO)

The United Nations System Document Ontology (UNDO), provides a formal representation of UN documents basic entities and their relationships. Each entity within the domain of UN can be defined in the documents and referenced through univocal identifiers (Internationalized Resource Identifier - IRI). While UNDO is specific for UN-related entities, other ontologies, thesauri (e.g. AGROVOC by FAO or UNBIS Thesaurus), or vocabularies (e.g. UNTERM) can be then referenced in AKN4UN documents, creating a true ecosystem of documents.

Sustainable Development Goals Taxonomy

In December 2017, an informal working group led by the Department of Economic and Social Affairs and the Secretariat of the United Nations High Level Committee on Management, and composed by technical experts from across the UN System, started working on a proposal to developed a system of Internationalized Resource Identifiers (IRIs) for Sustainable Development Goals, Targets, Indicators and related statistical series to support the development of a United Nations system-wide ecosystem of machine-readable data and documents based on Linked Open Data (LOD) principles. In order to ensure their fullest possible use, the Identifiers and a formal Statement of Adoption were presented at the second regular session of the UN System Chief Executives Board for Coordination (CEB), in November 2019. At the CEB, the Secretary-General invited all UN organizations to use them to map their SDG-related resources and sign the Statement.

UN Knowledge Extraction Tool

Adopting UNSIF was a first step towards a UN-wide ecosystem of machine-readable documents, to transform information hidden by traditional word processing into data interpretable by computers.

Countless innovative services can be built to foster collaboration and reduce costs in managing information across the system.

In 2020, HLCM worked with Microsoft to create a tool, based on Azure semantic technologies, trained to automatically extract information from UN General Assembly resolutions. The tool is open-source: every entity can adopt it and change it to fit their purposes.

Partnerships

• African Legal Institute
• European Parliament
• Italian Senate
• National Congress of Chile - Library
• General Assembly of Uruguay
• United States House of Representatives
• Brazilian Federal Senate

UN System Pilots

Links

• AKN4UN and UNDO Specs - GitHub
• OASIS webpage on Akoma Ntoso
• Visualizations and machine-readability at DGACM