Analysis Model of Customer Reviews Based on Neural Network

Reviews of customers affect the sales of e-commerce and positive customer reviews will boost store sales. Thereby, the emotional tendency of customer reviews is important to the evaluation of the business status of online store. In this paper, we use the Neuro-Linguistic Programmin algorithm to quantify the customer review, and then use the neural network to predict the emotional tendency of customer reviews by effectively analyzing the nonlinear relationship among affecting parameters. With the emotional tendency predication, customer reviews can help manage online store well. Simulation results show that our method can achieve high prediction accuracy. The accuracy rates of satisfaction prediction of the three commodities are 91.95%, 89.93%, 90.96%, respectively.