Sentiment analysis has recently become increasingly important with a massive increase in online content. It is associated with the analysis of textual data generated by social media that can be easily accessed, obtained, and analyzed. Just a few studies utilized sentiment analysis of social media using a machine learning approach. These studies focused more on sentiment analysis of Twitter tweets in the English language and did not pay more attention to other languages such as Arabic. The algorithms applied in the previous studies  were included  :Support Vector Machine Model, Neural Network Model ,Stochastic Gradient Descent model4-K nearest neighbors model, Naive Bayes model, Logistic Regression model, Ensemble learning Stacking Method (Random forest, Neural Network and KNN), Ensemble learning  Extreme Gradient Boosting Machine (XGBoost) .While the decision tree  (DT) model was used as a baseline classification comparison with previous best performing models This study proposes a machine learning model to analyze the Arabic tweets from Twitter. In this model, we apply Word2Vec for word embedding and Several algorithms have been applied in order to reach the highest accuracy in analyzing the sentiments of Twitter users, The best performing model was obtained in our data set. By getting the best hyperparameters This is when we use hyperparameter optimization, where we use Grid Search technology   The results show that applying word embedding with an ensemble XGBoost achieved good improvement on average of F1 score and the best of accuracy.