Optimizing E-commerce Recommendations with Doc2Vec and GRU

This paper explores the development of personalized recommendation engines for e-commerce platforms using end-to-end learning from user interactions. By analyzing user behaviors, such as purchased, added, or clicked products, personalized recommendations are generated using Doc2Vec and GRU algorithms. Users are grouped based on their interests, and tailored product suggestions are provided. The effectiveness of the recommendations is evaluated by optimizing long-term and short-term user intentions. This approach aims to enhance user satisfaction and maximize sales on e-commerce sites.