Analysis of promotional online shopping behavior based on machine learning

Abstract

Artificial intelligence's widespread use in e-commerce enables precise, personalized services and recommendations through deep data analysis, enhancing user experience and loyalty. Studying user behavior is essential for AI applications, providing businesses with valuable insights into customers' needs and behaviors, ultimately boosting sales efficiency and market competitiveness. This paper aims to investigate the application of artificial intelligence in the e-commerce domain, focusing on user behavior analysis and purchase behavior prediction. Using a user behavior dataset provided by the Tianchi platform, machine learning models, particularly random forest and logistic regression models, are constructed to predict user behavior in purchasing specific product categories. The results demonstrate that the random forest-based machine learning model can predict user purchase behavior effectively. Furthermore, the paper provides statistical data on user behavior and introduces related techniques such as feature engineering, performance metrics, and algorithms. The study highlights that artificial intelligence can offer personalized recommendations and services by analyzing users' historical behavior, interests, and preferences, optimizing merchants' product and service strategies, and ultimately enhancing sales efficiency and user satisfaction.