Unleashing the power of knowledge is imperative for enterprises looking for a competitive edge.
Everyone wants to capture knowledge, to connect everything that they know. However, turning data into knowledge is still very much an ongoing effort – though progress has been made, most data landscapes are far from mature. It’s time to connect data so it’s manageable and useful.
In this blog series, we’ll guide you from graph to knowledge graph – by starting small, gaining immediate value, and then expanding outward – the start of a short journey to unlimited insights.
Part 1: Data Trends and Challenges
Part 2: How a Graph Becomes a Knowledge Graph
The world of knowledge graphs, like knowledge itself, is multifaceted and broad, almost endless. In the final blog of our three-part series, we’ll discuss the two general categories of knowledge graphs: actioning knowledge graphs and decisioning knowledge graphs.
Actioning Knowledge Graphs for Data Management
Data management is an important use case for knowledge graphs. Born-digital companies like Lyft and Airbnb thrive on data, and enabling their data scientists to find the latest, freshest data is key to their success.
These companies and others have used knowledge graphs to create metadata hubs to capture data lineage: where their data came from, how it was transformed, and how it was cleansed. Knowledge graphs model complex data pipelines so that consumers and producers of data can be easily identified and new data sources can be integrated.
With a strong foundation regarding the provenance of data, you can take action on that data, confident of where it came from and who its producers and consumers are.
In addition to data management use cases, actioning knowledge graphs are used for personalization and recommendations. Actioning knowledge graphs bring together all data about customers, products, and more into a 360-degree view that drives a wide variety of actions such as identification of customers in danger of churning, as well as recommendations about the kind of offer that might persuade them to stay.
Decisioning Knowledge Graphs for Data Analytics
Knowledge graphs form the foundation of modern data and analytics. With data captured in a knowledge graph, you no longer need to guess at correlations: all the relationships inherent in your data are captured and stored. In this way, knowledge graphs represent a more faithful representation of data and enable you to unlock its predictive power.
With a decisioning knowledge graph, the ultimate goal is to make a better decision, whether that’s a human decision or an algorithmic decision. Those decisions can be supported in several ways.
Graph queries enable you to answer any question of your knowledge graph, at scale. Boston Scientific uses advanced queries to do root cause analysis and identify combinations of at-fault components that result in defects (an anti-recommendation of sorts).
Graph algorithms identify patterns in your data, such as the shortest path between two points or the most influential customers.
OrbitMI uses a decisioning knowledge graph to perform complex route planning for container ships. Using pathfinding algorithms, they plan maritime routes in less than a second. Furthermore, their knowledge graph backs their SaaS analytics offering. The knowledge graph’s impact is not only economic: greater efficiency in complex route planning reduced carbon emissions by 60,000 tons.
Graph queries and algorithms can also uncover predictive features for machine learning.
AstraZeneca uses graph algorithms and machine learning on its knowledge graph to identify patient journey archetypes and patterns. This research enables the company to identify influential touchpoints for early interventions to improve patient outcomes for illnesses like kidney disease.
Whether you’re looking to build an actioning knowledge graph or a decisioning one, Neo4j offers the most comprehensive knowledge graph on the market for data management, data analytics, and all the way to machine learning.
Get the White Paper