Bayesian Regularized Artificial Neural Network for Sentiment Classification of Amazon Customer Reviews

Author(s)	Mr. Abdul Hamid
Country	India
Abstract	Sentiment classification plays a vital role in extracting customer opinions from online reviews and supporting intelligent decision-making in e-commerce. Conventional machine learning techniques often experience reduced performance due to high-dimensional textual features and limited generalization capability. This paper proposes a Bayesian Regularized Artificial Neural Network (BRANN) for sentiment classification of Amazon customer reviews. The proposed approach integrates text preprocessing, TF–IDF feature representation, Bayesian regularization, and Levenberg–Marquardt optimization to improve classification accuracy while minimizing overfitting. The Bayesian learning strategy estimates probabilistic network parameters instead of deterministic weights, leading to improved model stability and prediction reliability. Experimental evaluation on the Amazon Customer Review dataset demonstrates an accuracy of 99.34%, precision of 99.18%, recall of 99.12%, and F1-score of 99.15% after 45 training epochs. Comparative analysis indicates that the proposed BRANN outperforms several conventional and deep learning models in terms of classification accuracy and convergence efficiency.
Keywords	Bayesian Regularization, Artificial Neural Network, Sentiment Classification, Opinion Mining, Amazon Reviews, TF–IDF.
Field	Engineering
Published In	Volume 7, Issue 4, July-August 2026
Published On	2026-07-27

About AIJFR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9898-79-39-59	editor@aijfr.com

Advanced International Journal for Research