Product dynamic pricing method based on transaction data and AI drive

Abstract

The invention discloses a product dynamic pricing method based on transaction data and AI driving, which is applied to product price prediction and aims at solving the problem that dynamic price prediction cannot be realized in the prior art. On the basis of the deep decision neural network, environmental parameters of the deep decision neural network are updated in real time in combination with dynamic sales volume prediction, and real-time feedback of the market trend is achieved; and meanwhile, a multi-layer full-connection neural network is adopted to model a key parameter relationship, such as a change relationship between sales volume and price, so that the expected value calculation accuracy is improved, and the effectiveness of the dynamic pricing method can be effectively improved.

Description

technical field [0001] The invention belongs to the field of data processing, and particularly relates to a product dynamic pricing technology. Background technique [0002] The timeliness and accuracy of quotations are the prerequisites for the success or failure of transactions. For product production and sales enterprises, timely and accurate quotation is particularly important, and dynamic pricing is the basis and premise of accurate and fast quotation. The research on product pricing method can speed up the speed of quotation and improve the accuracy of quotation, which is of great significance for enterprises to seize the opportunity and win the market. Therefore, it is necessary to study the dynamic pricing method that supports the collaborative integration of cross-domain and multi-subject business, and solve the problem that the fixed business process of the current traditional ERP is difficult to adapt to the dynamic changing needs of the production enterprise's b...

AI Technical Summary

Problems solved by technology

[0007] (1) Revenue-based dynamic pricing methods are basically based on traditional market forecasting models, which are not suitable for pricing under uncertain demand conditions

[0008] (2) The existing neural network-based intelligent forecasting method focuses on historical transaction data, and the collected information is basically historical price or trend information, without considering the interaction between price and demand At the same time, it does not take into account the fluctuation of demand in the competitive environment, so it is difficult to provide a reference value for the dynamic pricing of enterprises

Images