Demystifying Consumer Sentiment: An Accessible Approach to Analyzing Earbud Reviews for E-commerce Insights

Aim: This research seeks to provide an accessible method to sentiment analysis for individuals without any advance computer science backgrounds. To illustrate this, we are conducting an analysis of online product reviews for "Noise" brand earbuds to assess consumer sentiment and satisfaction. Study Design: This study used a mix of quantitative and qualitative research methods. The data was collected by web scraping all 5,078 customer reviews for every one of the 38 earbud models available on the official "Noise" e-commerce website. Methodology: The 5,078 reviews were analyzed using Microsoft Excel and Azure Machine Learning to classify each one as either positive, negative, or neutral. The results were then visualized in two ways: a pie chart showed the overall distribution of these sentiments, and a word cloud highlighted the most common words used in the reviews. Results: An analysis of 5,078 reviews showed that the overwhelming majority were positive. Specifically, 94% (4,756 reviews) were positive, while only a small fraction were negative (2% or 124 reviews) or neutral (4% or 198 reviews). A word cloud was also created to highlight the main themes and keywords from the customer feedback. Conclusion: The study successfully demonstrated the practicality of using accessible tools for sentiment analysis. The findings offer valuable insights into customer satisfaction with "Noise" earbuds, validating the efficacy of the methodology. Future research could explore diverse datasets and advanced machine learning techniques to expand these findings.