Building a slide deck, pitch, or presentation? Here are the big takeaways:
- Google Cloud is using its collection of big data tools to analyze NCAA data to form real-time predictions during the Final Four games.
- Historical data can be used to help organizations across industries better understand their audience and hows its needs shift.
The NCAA college basketball tournaments are nearly impossible to predict, but Google Cloud is using big data analytics to better understand how the teams will play. For the 2018 Final Four, Google is using its big data and machine learning tools to try to anticipate what might happen during a live game.
As noted in a blog post, Google used similar tools in the 2017 tournament to surface historical trends and statistics about each team. For 2018, though, Google wants to go a step further and try to predict, in real-time, how certain aspects of the game will play out—for example, how many three-pointers a given team will attempt in the second half.
If successful, Google's work with predictive analytics at the NCAA tournament could provide a template for how businesses can use this same kind of statistical analysis through big data to predict customer demand, supply chain activity, and shifts in business processes.
The Google Cloud team will be on-site during the final games, ingesting data and setting up the algorithms that power its predictions. Then, during halftime, those predictions will be shared in live TV ads.
SEE: Big data policy (Tech Pro Research)
There will only be a few minutes for the Google team to take the grokked data and turn it into a commercial. For this, Google will be working with a real-time rendering system built by Cloneless and Eleven Inc., it noted in the post.
"Before the end of halftime, we'll hand off our newly-created TV ad to CBS and Turner for airing on TBS right before the beginning of the second half," the post said. "This is likely the first time a company has used its own real-time predictive analytics to create ads during a live televised sporting event—wish us luck!"
Google has made it clear that they will not be trying to predict the winner. Rather, the goal of the experiment is to see how well the data science team can use big data and machine learning to make game-time predictions on the fly.
Raw data is ingested as JSON, XML, and CSV files coming from RESTful services or brought from an FTP, a separate post said. Then, data is analyzed through tools like Cloud Dataflow and BigQuery before predictions are visualized and modeled.
Due to the massive amount of statistical data produced in sports games, the industry is ripe for big data-field digital transformation. In addition to the work happening between Google and the NCAA, Amazon is also working with the NFL on "Next Gen Stats," and big data was also used heavily during the 2017 MLB World Series.
As big data continues to permeate every aspect of business, it may be professional and collegiate sports that lead the way for predictive analytics.
- 60 ways to get the most value from your big data initiatives (free PDF) (TechRepublic)
- Big Data 2018: Cloud storage becomes the de facto data lake (ZDNet)
- Predictive analytics: The smart person's guide (TechRepublic)
- Big data and digital transformation: How one enables the other (ZDNet)
- 5 steps to extracting big data gold (TechRepublic)
Conner Forrest has nothing to disclose. He doesn't hold investments in the technology companies he covers.
Conner Forrest is a Senior Editor for TechRepublic. He covers enterprise technology and is interested in the convergence of tech and culture.