LSTM Based Text Classification

Abstract

This paper will review the literature on various  methods and algorithms for analysing text sentiment.  This paper will go through different machine learning  algorithms for detecting sentiment in a text. Many of  the algorithms used here had drawbacks, such as taking  longer to train the model and using small datasets to  train, resulting in lower performance. The performance  of the model was improved by using LSTM, and it took  less time to train the model. When compared to LSTM,  many other approaches, such as RNN and CNN, are  inefficient. Various companies use Sentimental Analysis  to better understand their customers’ reactions to their   goods.