Research on the Cross-Border E-Commerce User Purchase Decision Model Based on Big Data Analysis

This study proposes a cross-border e-commerce user purchase decision optimization model based on big data technology and matrix decomposition. To better capture the user's purchase behavior, this study first comprehensively collects user behavior data, product information and external environment data, and uses big data technology to efficiently process and analyze these massive data. By constructing a matrix decomposition model based on the user-product rating matrix, the potential correlation between user preferences and product characteristics is deeply explored, to provide accurate personalized recommendations for e-commerce platforms. This recommendation system can optimize product recommendations and guide marketing decisions based on the user's historical purchase records, behavior patterns and product characteristics. Simulation results show that the recommendation system based on this optimization model has improved the accuracy of user purchase decisions by about 15% compared with traditional methods. At the same time, the conversion rate of the cross-border e-commerce platform increased by 10% and sales increased by 18%. These data prove the potential of this model in improving user experience, enhancing platform competitiveness and promoting sales growth.