Training Watson to make useful sense of that information was, as Baughman says, "extremely complex." First, IBM had to teach Watson to understand football by having the system "read" over six million articles about the sport. What were the beat writers who covered the quarterback's team on a regular basis seeing? What were the hardcore football analysts, who can better account for things like injury and off-the-field distractions, saying about the quarterback's matchup with a particular set of defensive backs? However, the network wasn't doing much to incorporate unstructured data. To wit: the network's predictive tools would note that a given quarterback threw and ran for a certain number of yards and produced a certain number of touchdowns on average each week, which would translate into a particular number of fantasy points.ĮSPN's subsequent fantasy performance predictions might then subtract a point or two from that average if that quarterback's next game was against a tough defense - as also measured by structured data. When network executives met up with some of Watson's developers in the spring of 2017, they discussed how ESPN was using a great deal of structured data to make its fantasy football predictions. But Syken says this partnership is different - rather than IBM creating data and then asking ESPN how they might use it, IBM instead went to ESPN and asked, "How can we help you?" IBM had provided assistance to the network in various ways in its golf and tennis coverage, Syken says, and Watson in particular had helped create AI-powered video highlight packages for both the Masters and the U.S. This wasn't Watson's first foray into sports, and it wasn't IBM's first team-up with ESPN. "But it's also applicable to legal and medical and financial trade-offs that people make every single day." "Being able to tap into both structured and unstructured data to allow you to make the most informed decision - it's applicable to fantasy sports, obviously," says Noah Syken, vice president of sports and entertainment partnerships at IBM. And fantasy football, those scientists say, is only one of many applications where AI's ability to read unstructured data could add a new level of sophistication to making decisions. Over the last decade, the analytics revolution has transformed sports, providing fans and fantasy team managers with seemingly endless statistics.īut how do you balance out all of the hard or structured data found in all those box scores, player sheets, and game logs with the game's wealth of unstructured data - the hundreds of thousands of words from online sources and social media that provide everything from locker room intel to injury analysis, all of which can help predict player performance? Here, IBM's data scientists figured, was where Watson could come in handy. He had help in the form of an unusually smart talent scout: IBM's Watson, which used artificial intelligence to make critical recommendations each week: predicting the best players to start games and the players to pick up off the waiver wire.įor Watson, fantasy football and its 60 million fans in the United States presented a huge opportunity. The IBM data scientist's fantasy football team ran the table, going undefeated in the regular season, putting up a 13-0 record. It wasn't just the Philadelphia Eagles and the New England Patriots who had great seasons. The result was a series of recommendations that produced an undefeated fantasy football team - and the promise of similar success stories for other businesses. An IBM data scientist asked Watson to study football: learning the language of the NFL and each of its more than 400 players, and fantasy football.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |