E-commerce Platform Risk Identification Using AHP Hierarchical Analysis and BPNN Neural Network

Artificial intelligence and machine learning are widely used in risk identification. Cross-border e-commerce platforms have increasingly become the best channel for global consumers to make purchases; however, with the surge in orders comes the risk of online purchases. Extensive quantitative data was collected through surveys and used for correlational studies that identify eight major risk factors, which are sales risk, cash/capital flow risk, market risk, customs clearance risk, shipping and delivery risk, environmental risk, logistics risk, and return risk. We then rated three online e-commerce platforms-Amazon, Ebay and Shopify-using AHP hierarchical analysis. The large number of survey responses collected, combined with the AHP scoring matrix we constructed, led to the conclusion that Shopify's online shopping risk is largest, followed by Ebay and Amazon. Based on this, we further establish and utilize BPNN neural network to predict the risk level of each website. The results show that the neural network trained in this paper has high accuracy and can be used for international e-commerce platform risk identification and evaluation.