Used Sailboat Price Prediction Based on Hybrid PSO-XGBoost Algorithm
Abstract
The price of luxury goods is often influenced by many factors, the characteristics of the luxury goods themselves, the market environment, the region where they are located and other factors that will make their prices fluctuate. On this basis, the price law of used luxury goods is more complicated. Used sailboats, as a typical category of used luxury goods, are loved by many merchants and customers. Since the prediction of its price is extremely complicated, there is a lack of hybrid models that can better deal with this problem in the past. To this end, this paper proposes a hybrid PSO-XGBoost algorithm based on the PSO algorithm and the XGBoost algorithm, and uses the proposed algorithm to predict the prices of used sailboats. Experiments show that the proposed hybrid PSO-XGBoost algorithm achieves R2 and MAPE of 0.802 and 16.913, respectively, on the test set, which has a better performance compared with the remaining five commonly used algorithms and can better predict the price of used sailboats.