<![CDATA[This study evaluates customer satisfaction in business and e-commerce using sentiment analysis based on Indonesian Lexicon and TextBlob. The method used in this study is an explorative quantitative approach with sentiment analysis techniques that compare the Lexicon and TextBlob methods in processing customer review data. The analysis results show the dominance of the neutral sentiment category, with Lexicon producing around 1400 neutral reviews, 1000 positive reviews, and less than 200 negative reviews, while TextBlob shows more than 2000 neutral reviews with less than 500 positive reviews and almost no negative reviews. These findings reveal that the Lexicon method is more sensitive in detecting positive sentiment than TextBlob, which tends to be conservative. The implication of this study is the importance of choosing the right sentiment analysis method to improve customer service strategies. With an accuracy score of 78.52%, precision of 68.11%, and F1-Score of 63.54%, this analysis provides practical insights into how companies can effectively interpret customer sentiment to improve service quality and overall customer satisfaction.]]>
Copyrights © 2024
