## A Knowledge Organization System for the United Nations Sustainable Development Goals > [!Abstract]- > This paper presents a formal knowledge organization system (KOS) to represent the United Nations Sustainable Development Goals (SDGs). The SDGs are a set of objectives adopted by all United Nations member states in 2015 to achieve a better and sustainable future. The developed KOS consists of an ontology that models the core elements of the Global SDG indicator framework, which currently includes 17 Goals, 169 Targets and 231 unique indicators, as well as more than 450 related statistical data series maintained by the global statistical community to monitor progress towards the SDGs, and of a dataset containing these elements. In addition to formalizing and establishing unique identifiers for the components of the SDGs and their indicator framework, the ontology includes mappings of each goal, target, indicator and data series to relevant terms and subjects in the United Nations Bibliographic Information System (UNBIS) and the EuroVoc vocabularies, thus facilitating multilingual semantic search and content linking. > [!Cite]- > Joshi, Amit, Luis Gonzalez Morales, Szymon Klarman, et al. “A Knowledge Organization System for the United Nations Sustainable Development Goals.” In _The Semantic Web_, edited by Ruben Verborgh, Katja Hose, Heiko Paulheim, et al., vol. 12731. Lecture Notes in Computer Science. Springer International Publishing, 2021. [https://doi.org/10.1007/978-3-030-77385-4_33](https://doi.org/10.1007/978-3-030-77385-4_33). > > [link](https://link.springer.com/10.1007/978-3-030-77385-4_33) [online](http://zotero.org/users/17587716/items/ADHLTGF7) [local](zotero://select/library/items/ADHLTGF7) [pdf](file://C:\Users\erikt\Zotero\storage\KCWCCV2F\Joshi%20et%20al.%20-%202021%20-%20A%20Knowledge%20Organization%20System%20for%20the%20United%20Nations%20Sustainable%20Development%20Goals.pdf) ## Notes %% begin notes %% Describes the LinkedSDG application and an ontology for representing the SDGs. The authors use spaCy to extract exact matches using the UNBIS Thesaurus and then graph traversal of their RDF graph to link from terms to SDG concepts (e.g., from "marsh" to "water"). Also describes the mechanics of (1) representing a knowledge system (e.g., ontology) and (2) linking to external ontologies (and how to link back to theirs). %% end notes %% %% begin annotations %% ### Imported: 2025-10-20 9:40 am United Nations General Assembly also adopted in 2017 a Global SDG indicator framework [12], developed by the Inter-Agency and Expert Group on SDG Indicators (IAEG-SDGs) While data availability is still a challenge for some of these indicators, there are already more than 450 statistical data series linked to most of these indicators in the Global SDG Indicator database, many of which are further disaggregated by sex, age, and other important dimensions. Data from the Global SDG database is made openly available to the public through various platforms, including an Open SDG Data Hub7 that provides web services with geo-referenced SDG indicator dataset, as well as an SDG API8 that can support third-party applications such as dashboards and data visualizations. The multilingual United Nations Bibliographic Information System (UNBIS) Thesaurus [10], created by the Dag Hammarskj ̈old Library, contains the terminology used in subject analysis of documents and other materials relevant to United Nations programme and activities. It is used as the subject authority and has been incorporated as the subject lexicon of the United Nations Official Document System. UN Environment Programme (UNEP) proposed to develop an SDG Interface Ontology (SDGIO) [3], focused on the formal specification and representation of the various meanings and usages of SDG-related terms and their interrelations9. Subsequently, UNEP led a working group to develop the SDGIO, which either created new content or coordinated the re-use of content from existing ontologies, applying best practices in ontology development from mature work of the Open Biological and Biomedical Ontology (OBO) Foundry and Library. Open Biological and Biomedical Ontology (OBO) Foundry and Library. the High-level Committee on Management (HLCM) of the United Nations System’s Chiefs Executives Board for Coordination adopted the UN Semantic Interoperability Framework (UNSIF) for normative and parliamentary documents, which includes Akoma Ntoso for the United Nations System (AKN4UN)10. Akoma Ntoso was originally developed in the context of an initiative by the UN Department of Economic and Social Affairs (UN DESA) to support the interchange and citation of documents among African parliaments and institutions and was subsequently formalized as an official OASIS standard. The group concluded a first draft of an SDG ontology as part of the SDG Knowledge Organization System (SDG KOS), consisting of a set of permanent Uniform Resource Identifiers (URIs) for the Goals, Targets and Indicators of the 2030 Agenda and their related statistical data series based on the SKOS model. The SDG ontology has been enriched with additional mappings to existing vocabularies and ontologies in order to facilitate content cataloguing, semantic search, and content linking. Specifically, each goal, target and indicator has been mapped with topics and concepts defined in UNBIS and EuroVoc19 thesauri. Identifiers have also been mapped to external ontologies like SDGIO and Wikidata. The UNBIS Thesaurus contains more than 7000 concepts across 18 domains and 143 micro-thesauri with lexicalizations in the six official languages of the UN, providing one of the most complete multilingual resources in the organization. EuroVoc is European Union’s multilingual and multidisciplinary thesaurus that contains more than 7000 terms, organized in 21 domains and 127 sub-domains. The ontology offers a web page for human consumption describing its content, resolving to the ontology file24 in case of requests for RDF data. VoID is an RDF vocabulary for describing linked datasets, which has become a W3C Interest Group Note27. VoID provides the policies for its publication and linking to the data [6] and also defines a protocol to publish dataset metadata alongside the actual data, making it possible for consumers to discover the dataset description just after encountering a resource in a dataset. Developed within the scope of the OntoLex W3C Community Group28, LIME is an extension of VoID for linguistic metadata. While being initially developed as the metadata module of the OntoLex-Lemon model29 [7], LIME intentionally provides descriptors that can be adapted to different scenarios (e.g. ontologies or thesauri being lexicalized, resources being onomasiologically or semasiologically conceived) and models adopted for the lexicalization work (rdfs:labels, SKOS or SKOS-XL terminological labels or Ontolex lexical entries). Finally, a list of subsets are then described in detail in the rest of the file. This list is mainly composed of void:Linksets, which are datasets consisting of a series of alignment triples between the described dataset and other target datasets, and of lime:Lexicalizations, the portions of the described dataset containing all the triples related to the (possibly multiple, as in this case) lexicalizations that are available for it. The SDGIO claims to be an Interface Ontology (IO) for SDGs, providing “a semantic bridge between 1) the Sustainable Development Goals, their targets, and indicators and 2) the large array of entities they refer to” . To achieve this objective, it “imports classes from numerous existing ontologies and maps to vocabularies such as GEMET to promote interoperability”. Among various connections, SDGIO strongly builds on multiple OBO Foundry ontologies “to help link data products to the SDGs”. Differently from the objectives of SDGIO, which is focused on its interfacing aspect and is based on a strong commitment to OBO ontologies, our mission was to build the official representation of SDGs and to place it under a largely interoperable, yet neutral, perspective. To this end, we represented SDG elements as ground objects, providing an ontology for describing their nature and mutual relationships, which builds in turn on top of the SKOS vocabulary, for purposes of visualization in SKOS-compliant consumers. The SDG ontology is part of an emerging system of SDG-related ontologies that aim to provide data inter-operability and a flexible interface for querying linkages across independent information systems. Mapping the identifiers described in these ontologies to each other and to external vocabularies allows SDG data to be clearly identified and found by semantic web agents for establishing further links and connections thereby facilitating knowledge discovery. A pilot application, LinkedSDG32, has been built to showcase the usefulness of adopting SDG KOS for extracting SDG related metadata from documents and establishing the connections among various SDGs. The application automatically extracts relevant SDG concepts mentioned in a given document using SDG KOS and provides their unified overview. The two key analytical techniques employed in LinkedSDG are taxonomybased term extraction and knowledge graph traversal. The term extraction mechanism, implemented using the spaCy library 34, scans the submitted document for all literal mentions of the relevant UNBIS and EuroVoc concept labels, based on the initially detected language of the document, and associates them with their respective concept identifiers. The traversal, performed using SPARQL via the underlying Apache Jena RDF store35, starts from these extracted concept identifiers, following to broader ones, to finally reach those connected directly to the elements of the SDG system via the dct:subject and skos:exactMatch predicates. Then, the algorithm traces the paths to broader SDG entities in the SDG KOS hierarchy. The application also provides access to the statistical data of the specific SDG series, which is represented as linked open statistical data using the RDF Data Cube vocabulary. In order to foster a holistic approach through coordinated policies and actions that bring together different levels of government and actors from all sectors of society, it is crucial to develop tools that facilitate the discovery and analysis of interlinkage across various global SDG indicators, as well as across other sources of data, information and knowledge maintained by different stakeholder groups. The SDG KOS is an attempt to provide stakeholders a means to publish the data using common terminologies and URIs centred around the SDG concepts, thus helping break information silos, promote synergies among communities, and enhance the semantic interoperability of different SDG-related data and information assets made available by various sectors of society. %% end annotations %% %% Import Date: 2025-10-20T09:40:27.039-06:00 %%