A Multimodel Transfer-Learning-Based Car Price Prediction Model with an Automatic Fuzzy Logic Parameter Optimizer

Cars are regarded as an indispensable means of transportation in Taiwan. Several studies have indicated that the automotive industry has witnessed remarkable advances and market used cars rapidly expanded. In this study, a price prediction system for BMW was developed. Nine parameters cars, including their model, registration year, transmission style, were analyzed. The data obtained then divided into three subsets. first subset to compare results each algorithm. predicted values produced by two algorithms with most satisfactory input fully connected neural network. second optimization algorithm modify number hidden layers network low, medium, high membership function (MF) achieve model optimization. Finally, third validation set during process. These subsets using k-fold cross-validation avoid overfitting selection bias. conclusion, combining optimal (i.e., random forest k-nearest neighbors) several gray wolf optimizer, multilayer perceptron, MF) successfully established. mean square error 0.0978, root-mean-square 0.3128, absolute 0.1903, coefficient determination 0.9249.