Table 1 Twitter sentiment analysis using DL techniques
From: Twitter sentiment analysis using hybrid gated attention recurrent network
Author & year | Methodology | Merits | Demerits |
|---|---|---|---|
Alharbi et al. 2019 [16] | CNN (Convolutional Neural Network) | The behavioural information of the user is included | Difficult to interpret the exact tweet from a group of tweets |
Tam et al. 2021 [17] | Hybrid CNN-BiLSTM (Convolution neural network and bidirectional long short-term memory) | The performance of the word embedding techniques is high | Lower classification and retrieval accuracy |
Chugh et al. 2021 [18] | DRNN (DeepRNN), SMO (Spider Monkey Optimization) and CSA (Crow Search Algorithm) | Provides better reviews to take effective decisions | Lower performance accuracy |
Alamoudi et al. 2021 [19] | Convolutional neural network (CNN), BERT and ALBERT models | Reduction in error rate | Occurrence of mislabelled reviews |
Tan et al. 2022 [20] | BERT approach (RoBERTa) with LSTM | Optimization is done using the word embedding technique | Lower classification accuracy |
Hasib et al. 2021 [21] | CNN (Convolutional neural network) and DNN | Collected data on the emotions of the airline consumers | Less number of tweets are used |
Guedes, G.P. 2020 [24] | UWS, SWS | Efficiency of proposed over the used dataset is found high | High error obtained |