Predict Click-Through Rates with Deep Interest Network Model in E-commerce Advertising

This paper proposes new methods to enhance click-through rate (CTR)prediction models using the Deep Interest Network (DIN) model, specificallyapplied to the advertising system of Alibaba's Taobao platform. Unliketraditional deep learning approaches, this research focuses on localized userbehavior activation for tailored ad targeting by leveraging extensive userbehavior data. Compared to traditional models, this method demonstratessuperior ability to handle diverse and dynamic user data, thereby improving theefficiency of ad systems and increasing revenue.