AN IMPROVED INTELLIGENT PRICING MODEL FOR RECYCLED MOBILE PHONES

Nowadays the market of recycled mobile phones has become so big that the autonomous pricing problem for the phones is attracting more and more attention. The pricing problem of recycled mobile phones is full of nonlinear mapping relationships and fuzzy concepts, which makes it difficult to accurately evaluate the value of recycled mobile phones. In order to improve pricing accuracy, an improved pricing model based on Fuzzy Neural Network with Momentum in Updating the Parameters (FNN-MUP) is proposed. First, we used principal component analysis to select the key features that affect the pricing of recycled mobile phones; second, we established a pricing model for recycled mobile phones based on fuzzy neural networks to achieve the mapping between key feature variables and cell phone price; finally, we applied momentum method to optimize the parameters of the fuzzy neural network to update the parameters. Experiments were conducted based on 1,200 pieces of data generated in real transactions for recycled mobile phones. The model was compared with other 6 pricing methods. The results show that the average relative error of this model is lower, and the prediction accuracy is significantly better than the contrasts.