Leveraging Big Data Analytics and Machine Learning Techniques for Sentiment Analysis of Amazon Product Reviews in Business Insights

Purpose: Satisfactory consumer feedback results from sentiment research which enables product quality enhancement. The research examines Amazon product review data through machine learning methods for sentiment analysis to extract important insights that improve customer experience. Materials and Methods: A Gradient Boost Classifier stands at the core of the proposed method which conducts sentiment analysis operations. The preliminary data treatment includes punctuation removal and stop word filtering followed by text tokenization. Feature extraction is performed using the Bag of Words (BoW) technique. The data is split into training and testing sets, and the models are evaluated using F1-score, recall, accuracy, and precision. Comparative analysis is conducted with Logistic Regression (LR), Naïve Bayes (NB), and Recursive Neural Network for Multiple Sentences (RNNMS). Findings: Among the tested models, the Gradient Boost Classifier consistently outperforms others, achieving a robust performance of 82% across all evaluation metrics. This highlights its superior classification capability in sentiment analysis tasks. Unique Contributions to Theory, Practice and Policy: While Gradient Boosting demonstrates high accuracy, future research could explore more advanced models and techniques, such as transformer-based architectures, to enhance sentiment classification across diverse product categories and address more nuanced sentiment patterns