Over 28,000 peer-reviewed scientific papers about graph-powered data science have been published in recent years. But access to the benefits of graph data science is no longer limited to scientists and those with deep pockets. The technology reveals contextual data connections, fuelling intelligent systems and enhancing machine learning predictions.
Google was among the first to use graph-based page ranking to revolutionise search engines. Now graph technology is experiencing exponential growth in usage. Interest in graph data science overlaps with AI and machine learning as companies seek to get the best insights from data.
Graph data science can reason about the ‘shape’ of the connected context for each piece of data through graph algorithms, enabling far superior machine learning modelling. Graph data science lets businesses make predictions in many diverse situations, from fraud detection to tracking customer or patient journeys. It helps companies learn from user journeys to present accurate recommendations for future purchases, supported by evidence from their buying history to build confidence in suggestions. Knowledge graphs are also being put to work to identify new associations between genes and diseases, discovering new drugs.
Analyst firm Gartner has identified graph data science as a key data and analytics technology trend. When asked about the use of AI and machine learning, 92% of companies said the plan to employ graph techniques within five years. Gartner believes a quarter of global Fortune 1000 companies will have built a skills base within three years and will be leveraging graph technologies as part of their data and analytics initiatives.
Graph technology in central government
Graph technology is being used at the top of government. Data scientists Felisia Loukou and Dr. Matthew Gregory deployed their first machine learning model with the help of graph technology to recommend content to GOV.UK users, based on the page they are visiting.
The scientists explain that their application learns continuous feature representations for the graph nodes, which can then be used for machine learning tasks, such as recommending content. They note that creating the necessary data infrastructure underpinning a model’s training and deployment is the most time-consuming part.
Graph tech powers marketing
Graph technology can contribute to identifying the right content even for casual visitors to websites. Ben Squire, senior data scientist at media and marketing services company Meredith, has used graph algorithms that transform billions of page views into millions of pseudonymous identifiers with rich browsing profiles. He says: “Providing relevant content to online users, even those who don’t authenticate, is essential to our business. Instead of ‘advertising in the dark,’ we now better understand our customers, which translates into significant revenue gains and better-served consumers.”
Supports the medical supply chain
Graph data science is also supporting the medical supply chain. Medical device manufacturer Boston Scientific is using graph data science to identify the causes of product failures.
Multiple teams, often in different countries, frequently work on the same problems in parallel. Yet engineers were having to resort to analysing their data in spreadsheets, leading to inconsistencies and, most importantly, difficulty finding the root causes of defects.
Boston Scientific set up a graph data model consisting of three nodes, with relationships that trace failures to parts and connect those to finished products. Analytical query times are much faster, helping to increase overall efficiency and streamline the entire analytical process. The company can now identify specific components that are more likely to fail. Another benefit is that because the graph data model is so simple, it’s easy to communicate to others. “Everyone involved with the project, from business stakeholders to technical implementers, is able to understand one another because they’re all speaking a common language,” confirms Eric Wespi, Data Scientist at the company.
Boosts health outcomes
In healthcare, New York-Presbyterian Hospital’s analytics team is using graph data science to track infections and take action to contain them. The hospital wanted to log every event, from the time a patient was admitted to all of the tests they undergo and their eventual release. Graph data science offers a flexible way to connect all the dimensions of an event – the what, when and where it happened.
The team created a ‘time’ and then a ‘space’ tree to model all the rooms patients could be treated in on-site. This initial model revealed a large number of inter-relationships. To build on this, an event entity was included to connect the time and location trees. The resulting data model allows the analytics team to analyse everything that happens in its facilities.
These examples are just a small taster of the possibilities of graph data science. Graph technology goes hand in hand with predictive AI and machine learning initiatives and is set to become a critical pillar of corporate data analytics.
The author is Lead Product Manager – Data Science at the world’s leading graph database