A Text Mining Analysis of Social Media Criticisms on Tuition Fee Policies at State Universities in Indonesia

This study aims to identify public criticism on social media related to the residential tuition policy imposed in state universities in Indonesia. Knowing the public's criticism of the single tuition policy on social media can identify critical issues, understand the public sentiment, and provide insights for policy improvement. The benefits of the process can be used to improve the quality of policies, strengthen relations with the public, increase student satisfaction and welfare, and contribute to academic knowledge about policy analysis using social media. The data came from public comments about tuition fees on Twitter, You'Tube, Instagram, and TikTok platforms. The dataset was labeled as neutral, negative, and positive sentiment in the first step using IndoBERT's pre-trained model. The result of a negative sentiment of 62.26% from the dataset was then used for topic analysis using the Latent Dirichlet Allocation algorithm, and a different number of topics was obtained on each platform with topic compositions for Twitter, YouTube, Instagram, and TikTok of 10, 5, 2, 10 topics respectively. The results of the synthesis from all platforms concluded that seven themes are the main issues discussed in public criticism related to tuition fees, where the top three themes are criticism of government policies that increase tuition fees (48,60/0), corruption cases occur in state universities (20,3 %) and students' concerns about continuing their studies when there is an increase in tuition fees (12,5<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup>).