Advances, Systems and Applications
Ref | Text Representation/Word Embedding | Model | Model description | Accuracy (%) |
---|---|---|---|---|
[43] | GloVe | Bi-LSTM -CNN | Multi Layered and Hybrid Deep Learning Model | 92.05 |
[44] | GloVe | 2 Bi-LSTM | Multi layer deep learning model | 87.46 |
[45] | GLoVe | Bi-GRU, Bi-LSTM | Hybrid deep learning model | 91.98 |
[46] | Word2Vec | CNN | Single layer neural network model | 89.47 |
[47] | Keras embed | CNN-LSTM | Hybrid deep learning model | 89.5 |
[48] | Keras embed | CNN-LSTM-CNN | Multi Layered Hybrid deep learning model | 89.02 |
[49] | TF-IDF, Count Vectorizer, Keras embed | LogR, SVM, MNB, CNN, RNN, LSTM | Single layer deep learning model and traditional Machine Learning models | 88 (RNN) |
[50] | Word2Vec | RNN | Single layer neural network model | 87 |
[51] | Word2Vec | CNN-LSTM | Hybrid deep learning model | 88.3 |
[52] | GloVe | LSTM-GRU | Hybrid deep learning model | 87.10 |
[53] | BOW | Multi Layer Perceptron | Single layer neural network model | 86.67 |
[54] | Word2Vec | Multi Layer Perceptron, CNN, LSTM and CNN-LSTM | Hybrid deep learning model and Single layer neural network model | 89.2 (CNN-LSTM) |
[55] | Unigram with TF-IDF | MNB, BNB, LogR, SVM, SGD, RF | Deep learning models and traditional Machine Learning models | 88.66 (RF) |
[56] | D-dimensional dense vector, n inputs, feature map of d × n in size | Hierarchical Conv-Nets | Numerous NLP tasks have been studied | 87.90 |