A Comparative Research On Sentiment Analysis of Tweets Using Deep and Machine Learning Models

Social media platforms are rich sources of user-generated material that are updated in real time like twitter and may offer important insights into public opinion. Taking advantage of the dynamic nature of user-generated information in real-time on social media sites like Twitter, this research uses a strict approach that includes data collection, sentiment categorization, feature extraction, and preprocessing. Sentiment analysis is guaranteed to have a representative sample when Twitter datasets are used. We combine machine learning models with deep learning models, such as CNN, RNN, FNN, and BiLSTM. We performed a comparative study of different models inorder to find the accuracy so that we can achieve the target that we are aimed to. This analysis provides future study for the topic twitter sentimental analysis for finding more effective techniques for more bird's eye categorization through comprehensive learning of machine learning and deep learning models.