Twitter Sentiment Analysis using RNN-LSTM

Abstract

    Sentimental analysis, one of the most famous utilization of Natural Language Processing, is broadly utilized, particularly as a piece of online media investigation for any space, be it a business, an ongoing film, or an item dispatch, to comprehend its gathering by individuals and what they consider it dependent on their assessments or sentiments. The main aspect of sentiment analysis is to examine an assemblage of text for understanding the sentiment communicated by it. Machine learning and deep learning models give the essential tools to insight examination in these challenges. This paper proposes LSTM neural network model which is prepared to anticipate textual information or opinions from twitter dataset and arrange those tweets into anger, joy, fear and sadness utilizing deep learning-based sentiment analysis strategies. Test results show that settings can assist us with performing supposition characterization amazingly preferably utilizing LSTM model over Machine learning classifiers.