The World Avatar
A knowledge-graph-based digital twin of the world
A knowledge graph is a network of data expressed as a directed graph, where the nodes of the graph are concepts or their instances (data items) and the edges of the graph are links between related concepts or instances. Knowledge graphs are often built using the principles of Linked Data. They provide a powerful means to host, query and traverse data, and to find and retrieve related information.
The World Avatar project, which is the flag-ship project in the C4T programme of CARES at the University of Cambridge, aims to create a digital ‘avatar’ of the real world. The digital world is composed of a dynamic knowledge graph that contains concepts and data that describe the world, and an ecosystem of autonomous computational agents that simulate the behaviour of the world and that update the concepts and data so that the digital world remains current in time. In this manner, it is proposed that it is possible to create digital twins that are able to describe the behavior of complex systems, and by doing so provide the ability to make data-driven decisions about how to optimise the systems.
A central tenet of the World Avatar is the requirement to share data so that all the computational agents experience a self-consistent view of the world, and the requirement that data and computational agents must be interoperable across different technical and social domains. This interoperability, defined in the sense of the ability of tools and systems to understand and use the functionalities of other tools, is considered to be one of the key aspects of Industry 4.0 and will become increasingly important as the complexity and choice of tools inevitably increases. For example, considerations raised by recent reports concerning the possible role of hydrogen in the future UK energy landscape span the electrical power system, the gas grid, the potential of biomass, solar and wind energy, the transport system, heating of commercial and domestic buildings, in addition to considerations relating to the acceptance and adoption of new technologies by the public.
The World Avatar represents information in a dynamic knowledge graph using technologies from the Semantic Web stack. Unlike a traditional database, the World Avatar contains an ecosystem of autonomous computational agents that continuously update it.

The main components of the dynamic knowledge graph are illustrated in the above figure. The dynamic knowledge graph contains concepts and instances that are linked to form a directed graph. The agents are described ontologically as part of the knowledge graph, and are able to perform actions on both concepts and instances. This confers versatility because it combines a design that enables agents to update and restructure the knowledge graph, with a structure that allows agents to discover and compose other agents simply by reading from and writing to the knowledge graph.
Recent Associated Preprints
308: Cross-Domain Flood Risk Assessment for Smart Cities using Dynamic Knowledge Graphs
Markus Hofmeister, George Brownbridge, Michael Hillman, Sebastian Mosbach, Jethro Akroyd, Kok Foong Lee, and Markus Kraft, Technical Report 308, c4e-Preprint Series, Cambridge, 2023.
307: Marie and BERT - A Knowledge Graph Embedding based Question Answering System for Chemistry
Xiaochi Zhou, Shaocong Zhang, Mehal Agarwal, Jethro Akroyd, Sebastian Mosbach, and Markus Kraft, Technical Report 307, c4e-Preprint Series, Cambridge, 2023.
306: A chemical species ontology for data integration and knowledge discovery
Laura Pascazio, Simon D. Rihm, Ali Naseri, Sebastian Mosbach, Jethro Akroyd, and Markus Kraft, Technical Report 306, c4e-Preprint Series, Cambridge, 2023.
305: British Wind Farm Battery Attachments: Curtailment Reduction vs Price Arbitrage
John Atherton, Jethro Akroyd, Feroz Farazi, Sebastian Mosbach, Mei Qi Lim, and Markus Kraft, Technical Report 305, c4e-Preprint Series, Cambridge, 2023.
Recent Associated Publications
Fully Automated Kinetic Models Extend our Understanding of Complex Reaction Mechanisms
Simon D. Rihm, Jiaru Bai, Laura Pascazio, and Markus Kraft, Chemie Ingenieur Technik 95(5), 740-748, (2023).
OntoPESScan: An Ontology for Potential Energy Surface Scans
Angiras Menon, Laura Pascazio, Daniel Nurkowski, Feroz Farazi, Sebastian Mosbach, Jethro Akroyd, and Markus Kraft, ACS Omega 8(2), 2462-2475, (2023).
Knowledge Engineering in Chemistry: From Expert Systems to Agents of Creation
Aleksandar Kondinski, Jiaru Bai, Sebastian Mosbach, Jethro Akroyd, and Markus Kraft, Accounts of Chemical Research 56(2), 128-139, (2023).
Automated Rational Design of Metal-Organic Polyhedra
Aleksandar Kondinski, Angiras Menon, Daniel Nurkowski, Feroz Farazi, Sebastian Mosbach, Jethro Akroyd, and Markus Kraft, Journal of the American Chemical Society 144(26), 11713-11728, (2022).
Funding
This research was partly funded by the National Research Foundation (NRF), Prime Minister's Office, Singapore under its Campus for Research Excellence and Technological Enterprise (CREATE) programme.