Diabetes is a chronic disease characterized by high blood sugar levels. Glucose is the main source of energy for human body cells, but in diabetics, this glucose cannot be used by the body. The International Diabetes Federation (IDF) organization in 2021 recorded 537 million adults (age 20 - 79 years) or 1 in 10 people living with diabetes worldwide. Indonesia is in fifth position with a population of 19.47 million people with diabetes. The government through Government Regulation Number 2 of 2018 and Minister of Health Regulation Number 4 of 2019 has determined that efforts to control diabetes mellitus are one of the minimum services that must be carried out by local governments. Every patient with diabetes mellitus will receive services at least once a month which includes measuring blood sugar levels, education, and pharmacological therapy as well as referrals if needed. This study discusses the classification of sentiment analysis of medical services in diabetes patients on Twitter using the naïve Bayes algorithm. Sentiment analysis was carried out to provide information on the level of public satisfaction with medical services for diabetes patients. The data used in this study were 11,704 tweets using four keywords, namely "diabetes services", "diabetes services", "blood sugar disease" and "blood sugar check". The results of the classification using Naïve Bayes obtained an accuracy value of 79% with a positive precision value of 73% and a negative precision value of 81%, a positive recall value of 61% and a negative recall value of 88% and a positive F1-score value of 66% and a negative F1-score value of 84 %.