E-Commerce Decision Model Based on Auto-Learning

The proposed model utilizes the information implied in the history of E-commerce negotiation to automatically mark the data to form the training samples, and apply the clues binary decision tree to automatically learn the samples to obtain the estimate of the opponent difference function. Then, an incremental decision-making problem is constituted through the combination of its own and the opponent's difference functions; and the dispersion algorithm is adopted to solve the optimization problem. The experimental results show that, the model still demonstrates relatively high efficiency and effectiveness under the condition of information confidentiality and no priori knowledge.