Discussing Data Science

A major theme emerged: Data analytics must empower individuals to take action in real time.

chart drawing photo

May 5th, 2015 was an exciting day for Big Data analytics. Intel hosted an event focused on data analytics, announcing the next generation of the Intel® Xeon® Processor E7 family and sharing an update on Cloudera one year after investing in the company.

https://youtu.be/kBxM2BCnoBk

At the event, I had the pleasure of hosting a panel discussion among three very interesting data science experts:

  • David Edwards, VP and Engineering Fellow at Cerner, a healthcare IT and electronic medical records company, has overseen the development of a Cloudera-based Big Data analytics system for patient medical data that has enabled the creation of a number of highly effective predictive models that have already saved the lives of hundreds of patients.
  • Don Fraynd, CEO of TeacherMatch, an analytics company that has developed models that correlate a broad variety of school teacher attributes with actual student performance measures to increase the effectiveness of the teacher hiring process. These models are used to identify the most promising candidates for each teaching position, given the individual circumstances of the teaching opportunity.
  • Andreas Weigend, Director of the Social Data Lab, professor at Stanford and UC Berkeley, and past Chief Scientist at Amazon, has been a leader in data science since before data science was a “thing.” His insights into measuring customer behavior and predicting how they make decisions has changed the way we experience the Internet.

My guests have all distinguished themselves by creating analytics solutions that provide actionable insights into individual human behavior in the areas of education, healthcare and retail.  Over the course of the discussion a major theme that emerged was that data analytics must empower individuals to take action in real time.

David described how Cerner’s algorithms are analyzing a variety of patient monitoring data in the hospital to identify patients who are going into septic shock, a life threatening toxic reaction to infection. “If you don’t close that loop and provide that immediate feedback in real time, it’s very difficult to change the outcome.”

Don explained how TeacherMatch is “using hot data, dashboards, and performance management practices in our schools to effect decisions in real time…What are the precursors to a student failing a course? What are the precursors to a student having a major trauma event?”

Andreas advanced the concept of a dashboard one step further and postulated that a solution analogous to a navigation system is what’s needed, because it can improve the quality of the data over time. “Instead of building complicated models, build incentives so that people share with you…I call this a data refinery…that takes data of the people, data by the people and makes it data to be useful for the people.”

Clearly, impactful analytics are as much about timeliness and responsivity as they are about data volume and variety, and they drive actions, not just insights.

In his final comments, David articulated one of my own goals for data science: “To make Big Data boring and uninteresting.” In other words, our goal is to make it commonplace for companies to utilize all of their data, both structured and unstructured, to provide better customer experiences, superior student performance or improved patient outcomes. As a data scientist, I can think of no better outcome for the work I do every day.

Thanks to our panelists and the audience for making this an engaging and informative event.

Related:
Download the CIO October 2016 Digital Magazine
Notice to our Readers
We're now using social media to take your comments and feedback. Learn more about this here.