Relational database systems are the de facto systems to implement data-intensive applications. Yet, many classes of applications require different data models, query languages, and/or computational capabilities from their underlying data systems. This 90-minute tutorial will cover three classes of these systems that are used by applications that process knowledge graphs:

Property graph databases, which adopt the property graph model and Cypher-like languages. This model arguably makes it easiest to bridge relational database users to the graph world.

Resource description framework (RDF) systems, which implement the RDF data model and the SPARQL language and provides automatic logical reasoning capabilities.

Datalog systems, which adopt the relational model but implement the logic-based Datalog language, allowing users to stay in the relational model while allowing them to do manual logical reasoning.

Semih Salihoğlu, Kuzu Inc, CEO

The tutorial provides a framework to evaluate the pros and cons of these different systems in terms of capabilities and performance. For each system, we cover an example application that uses a graph-based dataset. Examples involve path analysis in fraud and threat detection, logical reasoning to infer new facts in knowledge graphs, or using knowledge graph in retrieval augmented generation systems. The audience will learn about several open-sourced and/or commercial systems from each class of system, such as Relational AI, Kùzu, Neo4j, RDFox, and Ontotext.