Chetankumar Patel
E-commerce has fast become one of the major components of the worldwide online economy. It has positively affected the relationship between the business and the customer through the application of data and AI. At the same time, changing customer needs have turned understanding their preferences and offering them customized service into a major barrier to competitiveness. The study is focused on building an effective brand loyalty model that is able to not only support but also facilitate personalized marketing and customer service. The procedure is implemented using the E-Commerce Customer Dataset from Kaggle and involves a series of steps that include data cleaning, feature selection, and exploratory visualization through correlation and distribution analyses. Among the evaluated features, Satisfaction Score is found to be a crucial determinant of customer engagement and loyalty. The suggested Generative Adversarial Networks (GANs) model was trained, tested, and compared to several baseline machine learning (ML) methods such as Logistic Regression (LR), Random Forest (RF), and Naïve Bayes (NB). The GAN demonstrated its effectiveness in discovering sequential patterns and long-term relationships in behavioral data with 95.79% accuracy (ACC), 93.8% precision (PRE), 92% recall (REC), and 92% F1-score (F1) attained, surpassing all other models. The results show that GAN and other deep learning techniques outperform the standard models, making them useful tools for e-commerce platforms' predictive analytics, tailored marketing, and customer retention tactics.