A road map for accelerated discovery and learning
How to begin: steps to take now
Platforms, portals, projects, and organisations sharing content relating to climate action can already start preparing and contributing towards an integrated and unified view of heterogeneous knowledge relating to climate action by:
- Following existing good practice principles and standards where possible. Examples of such measures include “open government” principles, World Wide Web Consortium (W3C) standards, and FAIR principles (Wilkinson et al., 2016); and for taxonomies and ontologies, the Simple Knowledge Organization System (SKOS), and Web Ontology Language (OWL) standards.
- Sharing existing taxonomies and ontologies (both formal and informal) with one another to support widespread uptake and use, and to provide an overview of the terminology being used in different focus areas and within different websites. Where possible, joint mapping exercises can be used to (a) “link” existing terms (concepts) between the taxonomies, and (b) expand metadata associated with terms (such as their synonyms).
- Engaging experts to validate and improve taxonomies by adding missing terms, metadata, and information on how the terms in the taxonomies behave and relate to each other (their semantics) to produce ontologies. Metadata can include definitions from each community on the use of terms, as well as “scope notes” that describe context-specific use (and, importantly, excluded uses) to help non-experts understand and apply technical terms.
- Adopting and implementing shared taxonomies and ontologies within their websites to tag content with relevant key terms. A free to use tagging tool, such as the Climate Tagger, which also enables retrospective tagging of large datasets, could support this step. Importantly, the use of synonyms for key terms allows for individualisation of terms used by different websites.
- Developing application programming interfaces (APIs), to support interoperability and content sharing across websites. This is essential for crowd-sourcing, collating, and sharing pertinent knowledge.
- Promoting awareness of the added value and importance of IKM within and across institutions in supporting knowledge uptake, informing decisions, and enabling powerful analysis using AI approaches. This is particularly important for those in senior leadership and those in a position to direct investments towards IKM.
How to advance: steps to take over the medium and long terms
The activities advocated here involve the development of taxonomies and ontologies that follow SKOS and OWL standards. They require specialised expertise. As such, an initial step for all actors involved is to assess current capacity and literacy for this work, and to engage with IKM (specifically taxonomy and ontology) professionals to address capacity gaps. The goal is to use the shared taxonomies and ontologies for the development of a more comprehensive and sophisticated tool, a climate action knowledge graph. A knowledge graph is a model of a knowledge domain comprising concepts, classes, properties, relationships and entities covering multiple domains, various levels of granularity, and data from multiple sources.
The collaborating groups of actors (“actor groups”) must lead some of the steps by focussing on specific topics, sectors and frameworks (the “focus areas”). A wider community of actors must address other steps to contribute towards climate action. Here we highlight each step according to the participation and leadership required: whether led by actor groups; addressed as a community; or undertaken by a combination of the two.
Although presented here as a linear process, many of these activities can be undertaken in parallel and iteratively, enabling involvement of new actor groups at different stages. This process can emerge at different scales, for example within given subtopics of a wider agenda. Leadership and funding sufficient to direct and support collaborative action from the community are paramount. A European Union secretariat dedicated to enhancing IKM for climate change and sustainable development agendas could provide this.
- Collate and evaluate existing taxonomies and ontologies in relevant focus areas (topic, sector, policy framework).
- Collate all the different data, knowledge and information types that the shared taxonomy and ontology need to describe and relate.
- Conduct interviews and hold workshops with stakeholders to further explore the nature of content, terminologies and users’ information and knowledge needs, including the design of IKM systems and knowledge integration.
- Share, discuss and use outputs from steps 1-3 to explore significant overlaps in terminology and to establish components of a common ontology.
- Specify a set of (prioritised) core IKM activities that taxonomies, a common ontology, and the resulting overarching knowledge graph should support.
- Agree on standards for quality assurance, metadata, and governance of the taxonomies, common ontology, and knowledge graph, and make key decisions about their licensing and publishing.
- Agree on standards for the implementation and use of the shared taxonomies and common ontology to connect relevant content across websites, enable accurate clustering of knowledge for different decision-making contexts and ensure the linked data content pool is of sufficient quality to be useful to users.
- Develop a governance model that specifies how future changes and enrichments of taxonomies, common ontology and resulting knowledge graph will take place.
- Develop a common ontology framework to attribute characteristics to terms and describe the relationships between terms.
- Develop the focus area taxonomies and ontologies based on existing taxonomies and ontologies and their overlaps, the common ontology framework, the terminology used in that area, the content types that need to be described and the needs of stakeholders.
- Enrich and expand the taxonomies and ontologies through text analysis of documents, websites, and other content to identify new terms for integrating into the taxonomy.
- Add metadata to the focus area taxonomies to provide a rich base of information on the terms, including definitions and how they are used in different contexts .
- Analyse overlaps and, where appropriate, link the focus area taxonomies and ontologies to produce an integrated, shared climate action taxonomy and ontology.
- Implement the integrated taxonomy and ontology in knowledge management systems to produce a knowledge graph of climate action.
- Continue to enrich and expand the taxonomies, ontologies, and resulting overarching knowledge graph.
- Regularly test and evaluate the taxonomies, ontologies, and resulting knowledge graph and explore their potential to better support users, including through AI approaches and the development of “‘smart”, responsive IKM systems.
This road map provides an achievable pathway for transforming IKM to make the most of the vast array of climate-related knowledge already at hand but currently scattered across various sectors, communities of research and practice, organisations, governments, and private sectors. It enables the global climate change action community to build on, further develop and integrate existing work across these focus areas, to provide a global, integrated climate action data, information, and knowledge. These basic changes in organisational practices have transformative power. They can help people find and use needed knowledge. They can connect people and actions across the Web and across the world. They can empower policymakers, the private sector, the research community and society at large to collaborate in ways that allow marshal resources to address the defining global issues of our times.