An Intelligent Predictive Framework for Customer Satisfaction Analytics in E-Commerce

1 Vemuri Sasi Vardhan, 2 Yerraiahgari Venkata Abhinay Reddy, 3 RamiReddy Sathvik Reddy,4 Nimmagadda Chandra Sekhar

1,2,3 B. Tech Students, Department of CSE, RVR & JC College of Engineering, Chowdavaram, Guntur, A.P, India.

4Assistant Professor, Department of CSE, RVR & JC College of Engineering, Chowdavaram, Guntur, A.P, India,

E-mail:1 y22cs191@rvrjc.ac.in, 2 y22cs197@rvrjc.ac.in, 3 y22cs158@rvrjc.ac.in,  4 Nimmagadda65@gmail.com

Abstract:Vietnam’s fast-growing e-commerce market requires better tools to understand customer feedback. This study proposes a two-step framework combining deep learning (BERT, Bi-GRU) for sentiment analysis and machine learning (XGBoost) for predicting customer satisfaction. Using 10,021 reviews from major platforms (2015–2023), the models achieve over 70% sentiment accuracy and 80%+ satisfaction prediction accuracy, offering an effective way to enhance customer experience.