This research paper advances Sentiment Analysis and Aspect-Based Sentiment Analysis for e-commerce product reviews, delivering valuable insights for users and product creators. We extend beyond typical sentiment analysis by focusing on specific aspects of products such as features and qualities to provide a comprehensive analysis. Our methodology involves rigorous data preprocessing, including stemming, stop words removal, lemmatization, lower casing, contraction expansion, and tokenization, ensuring organized text data for efficient and accurate analysis. Central to our approach is the Aspect Term Extraction (ATE) step, which employs techniques such as POS Tagging, Noun Combination, Dependency Parsing, and Stop Word Removal to extract aspects from the text. These customized rules identify linguistic patterns and grammatical structures, revealing detailed facets of products. We fine-tune a BERT model for ABSA, processing the dataset to classify aspect sentiments as positive, neutral, or negative, and assigning a sentiment score ranging from –1 (extreme negativity) to +1 (extreme positivity). By leveraging advanced machine learning, we enhance the accuracy and depth of sentiment analysis, providing detailed insights into customer sentiments. This structured approach empowers businesses to extract actionable insights from extensive textual data and sets a new standard for sentiment analysis in digital retail.