What did positive and negative social media comments about Türkiye’s doctor appointment system focus on?

Details

Researchers analyzed public comments from X, Facebook, and YouTube to examine how people discussed the national doctor appointment system. They used a Turkish-language computer model to classify comments as positive, negative, or neutral, then analyzed topics separately within the positive and negative groups. On a random sample of 400 comments labeled by people, the sentiment model achieved 85.5% accuracy, so its classifications were not error-free.

Appointment unavailability and cancellation problems accounted for 86.29% of negative discussion. In positive discussion, smooth booking and system usability accounted for 64.95%, while institutional guidance and standard feedback messages accounted for 24.40%. Positive discussion therefore did not consist solely of personal reports of successful appointments.

Overall, 50.2% of comments were classified as positive, 29.4% as negative, and 20.4% as neutral. The researchers also identified a shift toward more positive average scores around November 2023. This observational analysis does not establish why the discussion changed or whether the service itself improved. It offers a complementary view of online feedback rather than a representative survey of everyone using the system.