Aspect level sentiment analysis/classification for text


What’s the best way to set up an aspect level classification model using the fast ai framework for text ?
For example, given a sentence which says “Movie1 is great but Movie2 is terrible”, the model should predict a positive sentiment if sentence, and “Movie1” is given as input; and a negative sentiment if the sentence and “Movie2” is given as input.

1 Like

I am also very much interested in this. @kkswam did you make any progress yet on your project?