{"id":25506,"date":"2024-07-31T13:18:09","date_gmt":"2024-07-31T11:18:09","guid":{"rendered":"https:\/\/www.uni.lu\/fstm-en\/?post_type=news&p=25506"},"modified":"2024-08-12T10:26:58","modified_gmt":"2024-08-12T08:26:58","slug":"olympic-handball-games-predicting-with-ai","status":"publish","type":"news","link":"https:\/\/www.uni.lu\/fstm-en\/news\/olympic-handball-games-predicting-with-ai\/","title":{"rendered":"Olympic Handball Games: predicting with AI"},"content":{"rendered":"\n
\n

Who will win the Olympic gold medal in handball this year? Mathematicians from the University of Luxembourg have developed an artificial intelligence model to predict and explain the results of the handball tournament at the 2024 Olympic Games. This is the first time a scientific method has been proposed to predict the final results.<\/p>\n\n\n\n

The Olympic Games are one of the main sports events in the world with 10,500 athletes from 203 national committees competing in 32 different sports. One of the team-based sports is handball, for which the Olympic tournament is considered as the most prestigious competition.<\/p>\n\n\n\n

\u201cOne application of my research is to develop an AI model to predict the results handball matches, with the focus on the main tournament for this year: the Olympic Games in Paris. The approach combines statistics, deep learning, and large language models to predict the score from a match. The model is based on historical data from past matches and considers factors such as team composition, match information, and statistical features to represent team strengths. Ultimately, the objective is to produce accurate predictions and also offer clear and understandable explanations to non-AI experts like handball coaches so it can help teams make informed decisions<\/em>\u201d, explains Florian Felice<\/a>, doctoral researcher at the University of Luxembourg and main author of the study.<\/p>\n\n\n\n

According to the predictions, France is considered the favourite team for both the men’s and women’s handball tournaments. When looking at the current results versus the predictions, the author of the analysis says:<\/p>\n\n\n\n

\u201cSo far <\/em>(as of July 31st<\/sup>), the predictions are pretty much aligned with the results. On the men\u2019s side, Croatia was a bit over-estimated but France\u2019s results are a big surprise with their unexpected performance on their first two matches. Though, predictions said they could not make it on their opening match against Denmark, their loss against Norway was a big surprise. It must come from big pressure of the event that the model can hardly account for. On the women\u2019s tournament, Netherlands seems to be stronger than expected while Spain seems to struggle getting their first win. But I have hope that both French teams will get to the gold, allez les bleu(e)s!<\/em>\u201d<\/p>\n\n\n\n

The final results to be delivered on August 10th<\/sup> for women and August 11th<\/sup> for men.<\/p>\n\n\n\n

Read the full publication:<\/p>\n\n\n\n