Authors: He Weiyi, Md. Yeasin Arafat
Abstract: Artificial intelligence (AI) has become an important technology for improving customer engagement, personalized marketing, and operational efficiency in cross-border e-commerce platforms. With the rapid growth of digital commerce and online consumer activities, understanding customer purchasing behavior through AI-driven analytics has become increasingly valuable for modern business optimization. This research presents an AI-based consumer behavior analysis framework for cross-border e-commerce optimization using multiple real-world datasets, including customer demographic information, shopping behavior data, social media advertising interactions, recommendation system data, and online purchase intention records. The study applies machine learning models, including Random Forest and XGBoost, to predict customer purchase decisions and analyze factors influencing online consumer behavior. Data preprocessing, feature engineering, exploratory data analysis, and classification techniques were implemented using Python-based analytics tools. Experimental results demonstrate that AI-driven models can effectively predict purchasing behavior and identify important factors affecting customer engagement and online purchase intention. The findings indicate that customer browsing behavior, social media advertising interaction, recommendation systems, and demographic characteristics significantly influence cross-border e-commerce purchasing decisions. This research contributes to the development of intelligent digital commerce systems by integrating AI analytics, consumer behavior analysis, and recommendation-based optimization strategies. The proposed framework provides practical insights for improving customer targeting, personalized marketing, and operational performance in international e-commerce environments.
DOI: https://doi.org/10.5281/zenodo.20489885
