Emotion Analysis for predicting the emotion labels using Machine Learning approaches

Advanced Web technologies used in this era, created huge amount of data provided to the internet operators Social Networking sites such as Facebook, Google, Twitter are platforms where people share and exchange their views, thoughts, discuss various things. The proposed method focuses on investigation of tweets data that show few positive or negative instances and also consider emotions in tweets data for sentiment analysis. Emotion Recognition and Detection from the content or documents is a neoteric area of research that is firmly related to Sentimental Analysis. Sentiment analysis and text emotion detection has achieved more popularity because of its huge efficient applications in almost every field such as psychology, political science, marketing, AI, human computer interaction etc. There are many Machine Learning techniques used for emotion detection and sentiment analysis. This work compares the different Machine Learning Algorithms Namely Logistic Regression (LR), Naïve Bayes and another comparison between RNN, GRU model and LSTM model. We use two datasets in this work, one for LR and NB, and another one for GRU model and LSTM model, then compare their accuracy results.