Research on Consumer Behavior Analysis and Recommendations for Cross-border E-commerce based on Big Data

At present, cross-border e-commerce is becoming an important transaction method for international trading. Therefore, the reasonable analysis for purchasing behaviors and precise recommendations are essential for the trading platforms. However, existing recommendation system relies on the collaborative filtering from users and ignore the extraordinary characteristic information from users including gender, hometown, and historical preferences. In this work, we initially determine the review scoring in consumer behavior, the influence of external factors such as time and geography on consumer behavior is analyzed, and the influence of time, space and other factors on the quality of merchants is eliminated. Subsequently, a training neural network is established to obtain the recommendations results for different consumers from the previous analysis module. Finally, we estimate our model in real-world commerce data and compare with existing recommendation algorithms. From our simulation results, we can observe that our proposed model can achieve the precise recommendations with reasonable time complexity.