Implement a Connected Inventory of enterprise data assets, based on a knowledge graph, to get business insights about the current status and trends, risk and opportunities, based on a holistic interrelated view of all enterprise assets.
Improve engagement, discoverability and personalized recommendations for Financial and Business Media, Market Intelligence and Investment Information Agencies,Science, Technology and Medicine Publishers, etc.
Turning Your Property Graph into a Robust Knowledge Graph
This is a step-by-step guide to transitioning from a proprietary labeled property graph (LPG) technology towards a standards based, semantics powered RDF knowledge graph (KG).
Migrating an LPG to a knowledge graph is a data migration and a technology migration project. When stated like this it sounds like a major endeavor, but it is probably the most straightforward project in that broad category you will ever undertake.
What is the Premise?
Starting with the observation that enterprises embarking on the knowledge graph journey via a labeled property graph backend quickly realize its shortcomings and face the necessity to migrate to a standardized semantics-based tech stack, we offer a step-by-step guide for performing precisely such an upgrade towards a true Enterprise Knowledge Graph. Each step is detailed with supporting examples and some theoretical background.
What does “Turning Your Property Graph into a Robust Knowledge Graph” include?
What is graph technology and why is there confusion between property graphs and knowledge graphs?
What is the blueprint for migrating an LPG to a knowledge graph?
What are the data modeling principles in LPGs and KGs – how the implicit schema in a property graph leads to an explicit semantic model? We address some of the fundamental differences between the two technologies, but as the Semantic Web stack is more expressive the ride is relatively smooth.
How to convert the raw graph data into an actual knowledge graph following an ontology? Every property graph data construct is directly mapped to an equivalent RDF construct. We also demonstrate some tooling for the automation of the ETL process.
How can queries in the Cypher query language be systematically converted to equivalent SPARQL queries? We also offer some examples and discussing philosophical differences between the two languages.
What Do You Need to Know in Advance?
While we assume only minimal exposure to the Semantic Web stack, this paper does not aim to be a tutorial on this technology. It can be read with only a superficial understanding of the Semantic Web to get a feel of what is possible and the scope of the effort, or it can be read as a practical guide to follow.
About the author: Borislav Iordanov is a Knowledge Engineering Consultant at Semantic Arts, Inc. He has over 25 years experience in the software industry with recurring focus on knowledge representation, model-driven development, distributed architectures and programming languages. His career includes co-founding several startups, driving innovation projects in several large organizations and more recently helping clients embrace a data-centric approach to information technology.
Want to address your enterprise data management challenges?
Fill out the form to get a free copy of this 24-page step-by-step guide!