Text bases Sentiment Analysis

Sentiment Analysis is one of the most growing fields of work in Natural Language Processing. Sentiment The analysis is used in determining the sentiment of any given text which can be positive, negative, and neutral. The proposed work includes the selection of databases, data acquisition, model building, and deploying the model. The databases selected were Airline tweets and Google Play App Reviews. Then the databases were pre-processed by techniques such as tokenization, lemmatization, and TF-IDF. Later, the pre-processed data are fed into the model. The LSTM (Long Short Term Memory) and BERT (Bidirectional Encoder Representations from Transformers) model was implemented. The LSTM model was trained and tested on Airline Tweets Sentiment Dataset whereas the BERT model was trained on Google Play App Reviews dataset and tested on a random user input text. Later, it was known the BERT model was performing better than the LSTM Model, so this model was selected and then deployed to form the Web Application programming interface (API).