Abstract—As Artificial Intelligence is developed to be more useful and used, expressing knowledge in a generalised, machine friendly way becomes important. In this article we start with an empty universe and from there set out to find the simplest possible structure to represent knowledge. We claim to find the atom of knowledge and give it the formal name Info Triple. Furthermore we use this structure to build a special kind of graph which we call Info Graph. The Info Graph can be used to model information context. We discuss the maths behind Info Graphs and compare it to existing standards. We compare the intended use of Info Graphs with the way graphs are currently used to model graph databases. Finally we conclude that the use of Info Graphs makes the models more generalised and therefore better suited for AI readers. Thus we believe our findings can serve as a valuable theoretical backbone for future knowledge base implementations.
Keywords—AI, graphs, info graph, info triple, information modelling, knowledge base
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Nicolaisen, U., "The Atom of Knowledge: Using Info Triples to Build Info Graphs - Short Version", Technical Disclosure Commons, (May 26, 2020)