Graph + AI World - Xandr Session

Graph + AI World: "Graph-Based Identity Resolution at Scale"

Graph + AI World Session - Recorded September 2020

One of the main drivers of the AT&T, WarnerMedia and Xandr success story is to combine their data assets to provide solutions for customers including the cross-device product offering, targeting the right users and content personalization. The data across these companies is not only large, containing billions of identifiers but also consists of various types of identifiers, for example, user accounts and cookies. Additionally, data from other third-party data providers is utilized to further extend our data coverage and provide a wider breadth of offerings to our clients. To this end, Xandr has built a multi-node, high-availability TigerGraph cluster to consolidate this data and store it in an identity graph. The Identity Resolution algorithm comprises of three components - label propagation, resolving large clusters and label persistence – which is run in a MapReduce fashion on the distributed identity graph.

About TigerGraph

TigerGraph is the only scalable graph database for the enterprise. Based on the industry’s first Native and Parallel Graph technology, TigerGraph unleashes the power of interconnected data, offering organizations deeper insights and better outcomes. TigerGraph fulfills the true promise and benefits of the graph platform by tackling the toughest data challenges in real time, no matter how large or complex the dataset. TigerGraph’s proven technology supports applications such as fraud detection, customer 360, MDM, IoT, AI and machine learning to make sense of ever-changing big data, and is used by customers including Amgen, China Mobile, Intuit, Wish and Zillow, along with some of the world’s largest healthcare, entertainment and financial institutions.

Graph + AI World: "Graph-Based Identity Resolution at Scale"

Graph + AI World Session - Recorded September 2020

Watch the Session

“TigerGraph is highly configurable ands can fit your specific use case. Whether you’re trying to get super fast query times or run heavy graph processing algorithms, setting the right configuration on your cluster will make all the difference.”

Xandr

About TigerGraph