Sentiment Analysis using Modified GRU
Aditya Agarwal, Prasanjit Dey, Sanjay Kumar · Proceedings of the 2022 Fourteenth International Conference on Contemporary Computing · 2022
Recently, everyone is becoming part of this digital world and spending time on social media like Twitter etc., posting their content. So a large number of heterogeneous Tweets have been coming each day. Due to heterogeneous structure, alone semantic Analysis can not classify and understand the sentences. To resolve this issue, the Deep Learning model plays an important role in understanding the meaning of content. In this paper, we proposed a modified Grated Recurrent Unit (GRU) for classification and understanding of the tweet. It consists of two gates update and a reset gate. The update gate knows how much memory to retain, and the reset gate knows how much to forget. We trained the proposed GRU model using the Sentiment140 dateset. Then experiments were conducted to find out the accuracy of the proposed model. Finally, we compared the accuracy of the proposed model with Long Short Terms Memory (LSTM) and Bi-directional Long short terms memory (BiLSTM). The experimental result shows that modified GRU outperformed the LSTM and BiLSTM models.