The Covid-19 virus is still very much causing symptoms to the community. Although not as severe as before, Covid-19 has not disappeared. Currently, people are very openly telling about themselves who are affected or have symptoms of Covid-19 on their various social media, one of which is Twitter. The dataset used for this research is data on Covid-19 symptoms from previous research by Sari (2022). TF-IDF weighting is used to calculate the number of occurrences of terms in the data, testing again using N-Gram as a word separation feature. Where the N-Gram is divided into 3 namely unigram, bigram and trigram. This research uses Random Forest and Support Vector Machine classification which tests the Linear, Radial Basic Function, and Polymonial kernels. The highest accuracy performance is obtained from testing Unigram with Support Vector Machine Linear kernel of 88.8%.