A Study on Medal Prediction Based on Sparrow Search Algorithm and MLP Neural Networks

In this study, a combination of Sparrow search algorithm and MLP neural network is innovatively utilized, and Logistic regression, EEMD integrated empirical modal decomposition, and DID double difference algorithms are integrated to focus on medal prediction, aiming at predicting the number of medals of different countries through accurate modeling. Sparrow search algorithm and MLP neural network are combined to predict the number of medals, successfully analyze the trend of the number of medals and the growth of the number of medals in several countries, and use Gaussian interval prediction to give the range of the number of medals of each country. Logistic regression is used to predict the first time winners, and the EEMD and DID analyses are used to explore the influence of other external factors on the number of medals. EEMD and DID analysis explores the impact of other external factors on the number of medals won. The advantage of this study lies in the efficient synergy of multiple algorithms. In data processing and model training, different algorithms give full play to their advantages in feature extraction, pattern recognition and trend prediction, and comprehensively consider the influence of various factors on the number of medals, which greatly improves the accuracy and reliability of the prediction results.