Social media is rapidly growing in Indonesia, replacing the role of conventional media in information dissemination. Indonesia ranks fifth in the world for the number of Twitter users. Twitter serves as a primary platform for users to express ideas, opinions, and criticism, but it also has negative impacts, one of which is body shaming. Body shaming is the act of making negative comments about a person's physique, such as "fat", "pug" or "cungkring", which often occurs on social media platforms, including Twitter. This study aims to analyze people's sentiments toward body shaming on Twitter using the Support Vector Machine (SVM) method. Datasets were collected through Twitter crawling techniques and then classified into positive, neutral, and negative sentiments. The confusion matrix evaluated the model, resulting in 66% accuracy, 69% precision, 66% recall, and 65% F1 score. The sentiment distribution shows that the positive class dominates with 1975 data (40.5%), followed by the neutral class with 1829 data (37.5%), and the negative class with 1071 data (22%). The results show that the SVM method is quite effective in classifying body shaming sentiment on Twitter. These findings can provide insights for developing sentiment detection algorithms and content moderation policies in social media.