<![CDATA[This study employs advanced sentiment analysis techniques to enhance the understanding of drug reviews, with a specific focus on TextBlob-based sentiment classification. As the accessibility of health products through pharmacies and online platforms continues to increase, individuals with limited health literacy are increasingly relying on user-generated feedback to inform their decision-making. By utilizing the TextBlob labelling method, this research categorizes user sentiments into positive, neutral, or negative, addressing the limitations inherent in traditional sentiment analysis approaches. The analysis is supported by an innovative model known as BERT, which effectively captures the emotional expression within textual data. The results indicate that the proposed approach consistently achieves an accuracy of 98% across training, validation, and testing phases, highlighting its strong performance in sentiment classification. This accomplishment underscores TextBlob’s ability to consistently and reliably assess user sentiment, thereby enriching the understanding of consumer perspectives in the pharmaceutical industry. The findings highlight the importance of effective sentiment analysis methods in healthcare, offering valuable insights for both consumers and stakeholders. Moreover, this study provides a foundation for future investigations focused on improving sentiment analysis methods across varied datasets, which will enhance the precision and applicability of classification results in different scenarios.]]>
Copyrights © 2025
