By Nik Vora, APAC VP at Neo4j
The exponential growth of Big Data has surpassed the point where traditional databases can manage it. Information isn’t only growing in volume but in complexity. Businesses are building vast repositories of data on their operations and their customers, with each entity having multiple points and layers of information. The problem is how to store, process and analyze this data in a meaningful and timely way.
Many organizations are now turning to graph data science to store data and generate insights. IDC estimates that big data technology and service-related spending will grow with a five-year CAGR of 14.6 per cent over the forecast period of 2020-25 in India. Accelerated demand is coming from a wide range of industries, including financial services, travel, retail, public sector and healthcare organizations, all wanting to solve complex problems.
Database platforms vs graph data science platforms
Representing today’s customer databases in a two dimensional table or spreadsheet is a very limited approach. Data can be stored and queried, but finding patterns among thousands of rows and cells isn’t an easy or immediate process. It’s extremely hard to connect different areas of data: for example not just who a customer is, but what they bought, how they bought, where they bought and why they bought.
Graph data science leverages the connections and relationships between billions and even trillions of data points. It lets the connected data “speak for itself”, such as running an unsupervised method of graph algorithm to find the signal in the noise. With a customer database it could show how the community of customers interacts, which could be useful information for segmentation.
For example, E-commerce company eBay used a knowledge graph to build an app for Google Assistant. It is coupled with natural language understanding and artificial intelligence to store, remember and learn from past interactions with shoppers. This helped them improve the ways shoppers search for the items they seek. When a user searches for a particular product, eBay App knows what details to ask about next, such as type, style, brand, budget or size. While accumulating this information by traversing through the graph, the application is also continuously checking inventory for the best match, thereby enhancing the entire real time decision making process.
In another example, Comcast used graph technology to derive insights into complex semantic and social relationships for their smart home devices.It helped create a scalable, flexible, multi-tenant user-profile service for extending personal information and relationships across multiple products. It models customers’ real-life relationships, and provides context so that the applications provide a more personalized experience for users.
A predictive, not reactive approach
In highly competitive markets like India, organizations need to stay one step ahead. For example, financial institutions typically address fraud when it has already happened. With graph data science, the suspicious connections between individuals and entities become visible and allow for much earlier intervention. A knowledge graph can identify chains and rings of linked individuals, scoring the quality, quantity and distance of one party’s relationships with suspicious entities.
When one fraud ring is identified, a similarity algorithm can also be used to identify other potential fraud rings across data. Once the kinds of patterns that predict a certain outcome are known, they can be used to generate much more accurate predictions in future.
A national finance ministry is using graph data science to map around 150,000 people, companies and documents, as well as approximately 750,000 relationships between these entities. If suspicious transactions are detected, the case is analyzed together with all relevant information and documents in the graph. Instead of taking a superficial look at relationships, legal experts can also uncover relationships only apparent at the second or third level.
Untangling APAC supply chains
One of the biggest disruptions in the past couple of years has been to supply chains – an issue intensely felt especially post the pandemic across sectors and businesses. This has only further emphasized the need to modernize supply chain management. Unraveling the extremely complex web of routes and participants to try to re-route tens of thousands of container ships crossing the oceans every day has been an immensely challenging task.
By nature, supply chain management is dynamic with many moving parts, and bottlenecks potentially occurring at any given point. But the volume and detail of data generated by traditional databases lacks real-time, accurate information processing capabilities.
Knowledge graphs are adept at mapping complex, inter-connected supply chains, and maintaining high performance even with vast volumes of data. Having an inherently relationship-centric approach makes them able to better manage, read and visualize their data. A graph database typically demonstrates 100 times faster query response speeds in contrast to a traditional SQL database.
Graph data science holds immense potential for organizations across the world and in regions like Asia that represent nearly 60 percent of the world’s population, organizations are harnessing the potential of big data through technologies such as graph data science and leveraging it to become world leaders in their respective sectors.