IBM has enhanced artificial intelligence (AI)-driven technology ahead of the return to live tennis. Powered by IBM Watson running on IBM Cloud, the company said the hybrid cloud applications for Wimbledon 2021 are based on the same AI and hybrid cloud technologies used with businesses across industries and around the world. Technologies developed by IBM provided for Wimbledon 2021 include enhancing a highly personalized fan experience, which will be available via Wimbledon.com and the Wimbledon apps. Some of the functionality also makes use of Red Hat OpenShift.
For the 2021 Wimbledon tournament, IBM said it had introduced three new fan experiences focusing on data, language, and automation. For the first time this year, the All England Lawn Tennis Club (AELTC) plans to offer IBM Power Rankings with Watson. This AI-powered analysis of player performance is presented as a leaderboard that updates every tournament day.
According to IBM, Power Rankings focuses on a player’s most recent match history, combining statistical analysis, the natural language processing of IBM Watson, and the IBM Cloud to analyze daily performance data, mine media commentary, and measure player momentum tournament-to-tournament and match-to-match.
IBM said Power Rankings analyses structured and unstructured data via an AI model built using IBM Watson Studio for Cloud Pak for Data. It said AutoAI in Watson Studio is used to help create AI models and generate candidate pipelines to reduce iterations for better predictions drastically.
Using this technology, IBM said the AELTC would offer fans predictive insights such as a pre-tournament view of players who have moved significantly up or down in the rankings and matches in which Power Rankings favor the lower-seeded player. Power Rankings also assign a win probability to both players ahead of the game.
During the 2020 US Open, IBM introduced Match Insights with Watson and extended this for Wimbledon with Pre-Match Insights. This provides what IBM describes as a “Watson fact sheet” for every main draw singles match. The AI-generated content will surface player insights to better inform fans of current player performance details ahead of each game.
IBM said the Pre-Match Insights functionality would present the IBM Power Rankings and generate a probability that a player will win before the match begins. The fact sheet provides an “In the Media” element, said IBM, which uses IBM Watson Discovery to create a custom news archive. Natural language processing (NLP) extracts key insights about each player from trusted news sources.
It also provides a “By the Numbers” section, which uses open-source natural language generation (NLG) to translate historical match statistics into player insights, combined with IBM Decision Optimisation, to select insights to display to the fan.
Alexandra Willis, head of communications, content, and digital at the AELTC, said: “We know how challenging tennis can be to follow and understand, especially for those who don’t follow the sport year-round. By creating these three new features – the IBM Power Rankings, the Pre-Match Insights, and the Personalised Recommendations and Highlights-, we hope to make it easier for fans to know who to follow and get more value from their experience with us.
“Partnering with IBM continues to be a critical part of our strategy to disrupt the traditional perceptions of Wimbledon through innovative and transformative uses of technology and deliver an incredible digital sports experience to fans worldwide.”
