Abstract
Currently, social networks have become the core of internet daily usage. Their popularity is increasing among the public users every day. Therefore, they can be considered as the main resource for gathering people's opinions and sentiments towards different topics. In this study, a comparison of three different machine learning algorithms used for sentiment analysis: Support Victor Machine, Random Forest Classification and K-mean Clustering is conducted. Unigrams and bigrams are used as features for all approaches. The results show that Support Vector Machine outperforms the other approaches.