The daily schedule of the Rio Olympic Games is wonderful. Table tennis, badminton and other items are eloquent. However, when the Chinese team played against international teams, the results seemed to be expected. Recently, for Chinese fans, the more exciting is of course the outstanding performance of Chinese women's volleyball player Holland in the finals. After a lapse of 12 years, the Chinese women's volleyball team once again advanced into the final of the Olympic Games. In the era of big data and artificial intelligence, everything seems less unexpected.
It is reported that Microsoft Xiao Bing has become the anchor of the Oriental Satellite TV Olympic News since August 12. In addition to continuously broadcasting the Olympics, it also predicts the results of the project. What is important is that Xiao Bing’s forecasting success rate of the results of the daily events in the program is also gradually improving.
As of press time, the Chinese women's volleyball team against the Serbian team, the two sides scored a fierce 2:1 battle. Just the day before, Xiao Bing announced the results of the forecast. Xiao Bing thinks that the Chinese women's volleyball team will defeat Serbia with a 57% chance of winning the gold medal.
According to reports, during the project forecasting process, Microsoft Xiao Bing collected the distribution of Olympic medals for each sports category since 1896, and the results of over 30,000 athletes in more than 50 countries and 28 major events. Create an initial event prediction model. At the same time, refer to each participating country's macroeconomic data such as per capita gross domestic product, population, per capita income, and other indicators, as well as political structure, climate, host country and other factors to establish economic forecasting model. In addition, referring to social platforms such as Facebook, Twitter, and Weibo, the users' inclination toward the public opinion of the Olympic Games and the degree of love of the athletes, etc., combined the above data, the regression model based on deep neural network obtained the final prediction results.