DIMENSIONAL SENTIMENT ANALYSIS USING REGIONAL CNN-ISTM MODEL

Dimensional sentiment analysis that means, continuous numerical values in multiple dimensions, such as valence arousal space (Degree of positive & negative sentiment)CNN(convolution neural network) it is used image recognition & classification algorithm at deep neural networks in the computer  visions.

Architecture of cnn-lstm model

 

  About model

By using word2vectoolkit,the vocabulary words are trained from corpus.

Cnn convert given text into r regions.

Effective featured should be extracted & pass convolutional layer &max pooling layers.

Convolutional layer:it is be extracted n gram features.

Maxpool layer:it apply max operation in each filters & elimates non maximal values ,finally reduces the computation.

Sequential layer: it integrates sequentially in each region vector into text vector.

Linear decoder: minimizing the mean squared error between the between the predicted output.

Conclusion

Dimensional sentiment analysis using regional CNN model can predict VA rating of the text

Leave a Comment

Your email address will not be published. Required fields are marked *