Thanks to Cornell natural language processing scholars, we know that female tennis pros field a greater number of unrelated tennis questions than their male counterparts. The Cornell team of Liye Fu, Cristian Danescu-Niculescu-Mizil and Lillian Lee used machine learning to look for language patterns within thousands of hours of in-game commentary and post-game interview transcripts. They found that roughly 70 percent of the questions unrelated to tennis were posed to female players.
Just as fascinating as the findings, however, are the team's methods.
Noted during one of pro tennis’s most celebrated tournaments, the New York Times spotlights the innovative thinking behind Cornell’s award-winning research and argues that algorithms and machine-learning are not mechanical; they are best used when combined with human creativity and ingenuity. The Cornell team’s work, the article notes, is a fine example of that:
“[T]he algorithm did not discover these biases on its own. This paper is cutting-edge research exactly because it required a spark of human intelligence. This was not a rote activity. It required months of work by some of the best researchers in the field of natural language processing.”
Read the team's 2016 International Joint Conference on Artificial Intelligence Best Paper here.