<![CDATA[This research aims to analyze public sentiment towards Starlink through tweets collected using the hashtag "starlink." The data crawling process was successful in collecting 1888 tweets. However, upon checking and processing the data, we reduced the number of valid and relevant tweets to 416. This reduction occurred due to duplicate data and the use of common keywords. We performed sentiment classification using the Naive Bayes model, yielding the following sentiment distribution: We classified 287 tweets (68.99%) as positive, 112 tweets (26.92%) as neutral, and 17 tweets (4.09%) as negative. The model performance evaluation shows good results with a recall of 0.80, precision of 0.90, F1 score of 0.83, and accuracy score of 0.80. The results of this study indicate that the majority of tweets related to Starlink have positive sentiments, indicating a generally favorable public perception of the service. A small proportion of tweets showed neutral and negative sentiments, which can provide valuable input for service improvement. The Naive Bayes model is able to classify sentiment with fairly high accuracy, making it one of the most effective tools for sentiment analysis.]]>
Copyrights © 2024
