The computational tools developed for big data have quickly begun to change the world, and it is essential that our students learn how to drive that transformation. Of course, you can only be in the driver’s seat if you know how to use the algorithms and where you want to go. Only engagement in real-world problems provides both the use case and the deeper understanding of what the big data transformation means.
Our DASH Co-Directors – Dan Price and Peggy Lindner – have been offering research opportunities with community non-profits and data visualization since 2014, and this year they have stepped up to a new challenge. In memory of a much-loved friend and colleague, we have begun the George “Trey” Pharis Memorial Fellowship Program, and in the inaugural year we will be using hypergraphs to model community health.
Hypergraphs are often used in machine learning and artificial intelligence to represent complex causal systems. They are an extension of the more common idea of a network graph – like the social networks familiar from online platforms like Facebook – but instead of just having single relations with others as friends, you can have many-layered and complex relations. It’s more like sets of overlapping Venn diagrams than a connect the dot.
Most importantly, you can use hypergraphs to represent the ways intentionally directed activities support the relationships, making the connections themselves stronger and not just doing something with one person at a time. This is extremely important in the world of community health, where it’s been very hard to show the value of community relationships. And if we can show how to place a value on the relationships, and what it means to demonstrate that community has been made stronger, we can help identify existing strengths in the community and model strategies and interventions with community strengths already in mind. In other words, we can do what we set out to do, and have our students drive the transformation through engagement and the hands-on use of data analytics.