Screen Shot 2018-11-13 at 11.00.07 AM


Finding Hubs of Influence - Implementing PageRank with a Native Parallel Graph Database

RECORDED October 30, 2018
This major graph analytics algorithm needs little introduction, as Google has made it a household term: PageRank. Here we are looking for the most influential or popular members of a particular group - whether it's a web page that is referred or linked to by the highest number of pages, subscribers driving the highest volume of calls in a network, doctors driving the maximum number of referrals for a particular condition such as diabetes, cancer or opioid addiction, or social and media influencers who are driving the largest number of referrals for an eCommerce, digital wallet app or ride sharing service.

In this Graph Gurus episode, we use the TigerGraph Developer Edition to build a solution using the “PageRank” algorithm, from the newly released TigerGraph algorithm library. In this episode, we will:

    • Explain the “PageRank” graph analytics algorithm and review multiple use cases for industries including telecom, government, healthcare, eCommerce, payments and ride sharing service.
    • Present a graph solution for a real-life business use case to perform PageRank using TigerGraph and GSQL.
    • Consider benchmark comparing performance of TigerGraph and Neo4j for PageRank algorithm for 500 GB dataset.


Dr. Victor Lee, Director Product Management