Sentiment analysis is one among the distinguished fields of knowledge and pattern mining that deals with the identification and analysis of sentiment within the text. The main challenges in sentiment analysis are word ambiguity and multi polarity. The problem of word ambiguity is to define polarity because the polarity for words is context dependent. The tweets are initially preprocessed. The preprocessing includes the removal of stop words, and lower case conversion. The tweets are then passed to the feature extraction techniques. Then the data is splitted as training and testing data. The trained data is passed to the different machine learning algorithm like Naive Bayes. Support Vector machine, Random forest, and Decision Tree and k-NN algorithm. The accuracy obtained using the Naive Bayes. Support Vector machine, random forest, and Decision Tree, k-NN and Logistic regression algorithm is 80%, 77%, 72%, 61% ,56% and 78%. The naïve bayes algorithm has achieved a better accuracy when compared to the other algorithm.