The transaction price prediction of second-hand cars based on model fusion

Abstract

With the gradual improvement of the second-hand car trading market, buyers and sellers are increasingly demanding a more orderly market, and price forecasting plays an important role in regulating the market order. On one hand, buyers hope to buy second-hand cars with better quality at a lower price. On the other hand, the seller wants to know what the buyer's demand is, so as to better sell second-hand cars in the market. Therefore, the reasonable prediction of price plays an important role in regulating the buying and selling market. At present, some machine learning models, such as neural networks, tree models are very mature in regression prediction technology. By using the above algorithm for modeling, the model can play the function of reasonably predicting prices, so as to meet the needs of both buyers and sellers. Therefore, in this paper, the author predicts the second-hand car transaction price through modeling around some machine learning algorithms, and analyses the performance of the model.