we proposed a emotion recognition system where it recognize the emotions in tweets. As emotions play a vital role in our lives. As we can see that many people use social media where they use the platform for many purposes, some of them tweet in a good way and some of them in a bullying way. Emotion and opinions of different people can be carried out on tweets to analyze public opinion on a news and social events that are taking place in present society. In this project, by using machine learning algorithms we have implemented emotion recognition by classifying tweets as positive and negative. By recognizing these positive and negative tweets we can identify people emotions where we can reduce the forged statements. Initially we have divided our dataset into train and test dataset, where it is used to train the model and by comparing the train data with the test data, the model recognizes the emotions in tweets. By using svm and naïve bayes algorithms we classify the text based on twitter into different emotions and predicted emojis like love, fear, anger, sadness, joy. Based on the performance analysis we predicted optimal result with 79% accuracy and 81% F1 score

LR-SGD, Emotion, Machine Learning models, TF, TF-IDF