OECD confirms deal to use AI-based UNSILO Classify for Sustainable Development Goals collections

(26 Sep 2019)  UNSILO and the Organisation for Economic Co-operation and Development (OECD) this week confirmed a multi-year partnership using UNSILO AI tools. The partnership, which was
successfully trialled last year, is to improve content discovery and collection-building of
thousands of reports, papers and data available on the OECD’s iLibrary publishing platform,
using UNSILO’s artificial intelligence technology. Specifically, the tool is being used to tag not
only the OECD’s iLibrary collection, but also IGO partner collections according to the 17 United
Nations Sustainable Development Goals (SDGs). After extensive trials, the beta release of the
SDG Pathfinder tool was launched at the May 2019 OECD Forum. After further testing, a newly
designed SDG Pathfinder will be released this coming November at the EU Digital Summit
and previewed at the October Frankfurt Book Fair at the UNSILO stand.

Underlying the SDG Pathfinder, which identifies content relating to each of the 17 SDGs, is the
UNSILO concept extraction engine, which uses semantic tools and machine learning to
extract and analyze key concepts from documents. The UNSILO concept extraction works in
any subject domain, and does not require a pre-existing taxonomy or classification system.
For OECD, concept extraction is manipulated by the OECD team using UNSILO Classify, a
desktop application that automatically creates thematic subsets of existing collections
around any broad or narrow subject. Classify provides a simple tool to create and maintain
new and innovative content packages that can support communication and dissemination
activities, without losing control of the process. The team of in-house subject matter experts
remain in full control of the degree of automation of the subject extraction, thereby reducing
the cost and time of manually creating collections of content.

Classify is used to identify relevant content across the iLibrary digital publishing platform,
pooling analytical content, published from 2010 onwards, from six international organisations,
as well as statistical content when available: International Telecommunication Union , Nordic
Co-operation , Organisation for Economic Co-operation and Development , The
Commonwealth , United Nations , and the World Trade Organization . “The UNSILO tool has
enabled us to deliver on our goals to innovate and maximise content delivery for users,”
commented Janine Treves, Digital Managing Editor, OECD Public Affairs and
Communications Directorate. “We needed to be able to classify thousands of documents
from a range of publishing publishers, and to do so as fast as they are published. We look
forward to furthering our partnership with UNSILO tools to keep innovating for better user

Thomas Laursen, Chief Executive Officer at UNSILO, stated: “We are extremely excited that
our collaboration with the OECD publishing team has been so successful, and we look
forward to further content initiatives with the team.”