IBM Uses Bluemix to Bring Watson to the US Open

 
 
By Darryl K. Taft  |  Posted 2016-09-04 Print this article Print
 
 
 
 
 
 
 
IBM Watson


Among the digital platforms IBM delivers for the USTA is a content management system that enables publishing to the USOpen.org website and all the organization's digital platforms in real time. New this year for the video process—specifically for on-demand video—IBM is using the Watson speech-to-text API available on the Bluemix platform to listen to videos and create subtitles automatically for the videos, as well as produce transcripts of the videos. All that data is entered into the content management system automatically. The USTA editorial team can edit and refine that text and then publish it to the platform, said Kent. And any changes they make are then sent back to Watson so it can learn from those adaptations.

Additionally, all the photos taken by the USTA photography team are all analyzed by Watson, and players and celebrities are automatically identified and meta-tagged so the USTA will have search capabilities within the publishing environment. So, as users are looking for photos of various players or celebrities to build galleries or tell stories, they will have quicker access to some of that content, Kent said. This uses the Watson Visual Recognition API, which also is available on Bluemix.

"Each year, IBM designs, develops and delivers a compelling digital experience that enables us to engage, entertain and inform our fans in entirely new ways, while transforming how they encounter and enjoy the tournament," said Kirsten Corio, managing director of Ticket Sales and Digital Strategy for the USTA, in a statement.

Kent said, for IBM, the goal is simple: to continually create experiences for tennis fans that are compelling and interactive. And at the heart of that is data. IBM collects all the data around a point and the chair umpire system collects the outcome of that point, such as whether there's foot fault involved. The courtside statistician collects additional data on whether that point was a winner or an unforced error. Radar and serve speed also are captured, and IBM also has access to the official line calling system for player and ball position data.

IBM sends that data around the grounds to all the scoreboards, to the media and to all the digital platforms. Last year saw more than 16 million new unique users of the digital platforms over the course of the two-week period of the US Open.

One of IBM's premier offerings for scoring is its SlamTracker app, now available on mobile devices.

"This is the first time SlamTracker is available on mobile here at the US Open, since we are starting to see a shift in user behavior from desktop to mobile," Kent said. "Last year mobile eclipsed desktop significantly, so we, like many businesses, are focused on the mobile experience. And SlamTracker allows for the real-time, point-by-point scoring, but it also acts much like a commentator after every point concludes—where it will act much like a commentator and offer some perspective on what occurred during that point."

SlamTracker also has a stage item in which IBM can offer some kind of insight. For instance, if a player hits an ace, IBM could put up a graphic of how many aces the player has or something relevant to the point to make it more informative for the person following along online.

Also, new to SlamTracker this year is some additional analytics, Kent said.

"We've done some analytics around something we refer to as ‘pressure situations,'" he said. "So, for example, when a player is down Love-40 or down in a match—two sets for a men's match—we've analyzed the last eight or nine years of Grand Slam tennis data to understand how that player performs in that particular situation. And the analytics are done on a Bluemix platform using Apache Spark."

Kent noted that at the core of SlamTracker's predictive analytics technology is its "Keys to the Match" feature, which analyzes eight years of Grand Slam tennis data comprised of 41 million data points.

"Watson is revolutionizing the way fans can navigate the tournament this year," said Noah Syken, vice president of global sponsorships and client executive programs at IBM, in a statement. "By tapping into unstructured data, Watson is enabling us to extract and apply insights that can improve how people engage with technology on-site, making their experiences more meaningful and natural."



 
 
 
 
 
 
 
 
 
 
 
 
 

Submit a Comment

Loading Comments...
 
Manage your Newsletters: Login   Register My Newsletters























 
 
 
 
 
 
 
 
 
 
 
Rocket Fuel