Knowledge graph using resource description framework and connectionist theory
Ravi Lourdusamy, Xavierlal J. Mattam · Journal of Physics Conference Series · 2020
Abstract Interest in Knowledge Graph has peeked these years. The use of RDF triples for the construction of knowledge graph has been limited by the fact that the vector embedding of RDF to make it machine readable has been constraining factor. This article presents the use of weighted RDF as a vector embedding of RDF that could be used with Bayesian networks in Graph Neural Networks. The vector embedding helps in representing the weights on RDF that could be obtained using Bayes theorem of assessing probability. The resulting weighted RDF could be used in many applications. The use of it in a Clinical Decision Support System is given as a simple illustration together with the calculation of the weights using Bayes theorem. The weighted RDF in Graph Neural Network will represent the knowledge graph using RDF and connectionist theory.