Research on Sailboat Price Prediction Based on LS-SVM Prediction Algorithm and XGBoost Algorithm

This paper investigates the prediction of sailboat prices using the LS-SVM prediction algorithm and the XGBoost algorithm. The study involves training prediction models, determining coefficients of determination, and calculating the mean absolute error (MAE) for both LS-SVM and SVM algorithms, resulting in values of 0.99143, 0.89572, and 0.94512, respectively. The LS-SVM model demonstrated the highest predictive accuracy. Additionally, a regional impact model incorporating the XGBoost algorithm was introduced. Data from each region was utilized to predict the performance of second-hand sailboats, serving as the training set for the model. The study culminated in the derivation of the solution model through the XGBoost algorithm.