Research On Intelligent Evaluation Method Of Used Car Price And Prototype System

With the rapid development of Chinese economy,the scale of used car market is increasing.In recent years,the consumption concept of consumers has changed.People gradually accept the purchase and use of used cars,so used cars have a good acceptance.The traditional evaluation method of used car price has complicated steps and strong subjectivity of evaluation results,which gradually becomes relatively backward.In order to provide a scientific price reference for used car transactions,and improve the accuracy of used car price evaluation methods,and promote the healthy development of used car market in China,the following contents are researched in this paper.Firstly,a used car price evaluation model based on MIV-GA-BP neural network is established.Since the used car has the characteristic of " one car,one condition,one price",the influence factors of the used car price are studied and analyzed,and 18 characteristics of the used car are selected from the micro factors to establish the BP neural network model.Based on the eigenvalues of each car as input,the neural network model is used to evaluate the used car price.In order to avoid BP neural network training falling into local extreme value,the initial weight and threshold of BP neural network model were optimized by using genetic algorithm,so the GA-BP neural network model was obtained.Then MIV variable screening method is used to simplify the input of the GA-BP neural network model to get the MIV-GABP neural network model.By comparing the correlation coefficient R and the mean square error of the models,it is verified that the MIV-GA-BP neural network model in this paper has higher accuracy than the traditional BP neural network model.Secondly,a kind of used car price evaluation method based on AHP Case-Based Reasoning is proposed.The MIV-GA-BP neural network model established in this paper is used to evaluate the price of the used car without accident,while the used car price evaluation method based on AHP Case-Based Reasoning considers the accident situation of the used car.The used car price evaluation method based on AHP Case-Based Reasoning is used to solve the current used car price evaluation according to the actual used car cases in the history of transaction.For this method,a case retrieval strategy based on Euclidean distance and triangular fuzzy number is designed,and the Analytic Hierarchy Process(AHP)is used to calculate the weight of the attributes of used car cases.Finally,an intelligent evaluation prototype system of used car price is designed and developed.In order to apply the MIV-GA-BP neural network model and the used car price evaluation method based on AHP Case-Based Reasoning to the actual used car price evaluation,as well as to improve the information management level of small and medium-sized used car enterprises,an intelligent evaluation prototype system of used car price based on B/S(Browser/Server)framework is designed and developed in Py Charm development environment.The My SQL database is used to store multiple related data tables and this prototype system has used car price evaluation module,sales management module and inventory management module,etc.This prototype system can not only be used to evaluate the price of used cars,but also can be used to manage various information in the operation process of small and mediumsized used car enterprises.