Due to the widespread nature of online shopping and the low barrier for uploading a message, sentiments or opinions expressed there to give the most up-to-date and comprehensive information possible. The field of sentiment analysis is becoming increasingly important, yet there is currently no comprehensive research on product reviews. In classifying the reviews of product reviews, it is crucial to (1) detail the obstacles that still need to be overcome, (2) highlight the most significant developments to date, and (3) assess how far it's come over the years. This research article proposes a novel classifier, namely Renewed Genetic Algorithm (RGA) to effectively classify the reviews of products on the Amazon Online Shopping Website. RGA concentrates more on local search for accurately classifying product reviews. Improved classification accuracy is achieved by basing classification on fitness computation. This research employed a review dataset consisting of 4 distinct product domains from the Amazon Online Shopping Website to compare how well RGA performs compared to existing classifiers. When compared to standard classifiers, the results favor the proposed classifier RGA. The renewed genetic algorithm (RGAC) has achieved up to 89% Precision, 86.8% F1-Measure, 86.4% Accuracy, and an MCC of 72.9% across Amazon review domains. These results represent gains of 10–25% over competitive baselines, confirming the method’s superiority in cross-domain sentiment classification.