I am beginner in natural language processing and my goal is to find a way to score sentences based on their emotional tension. More specifically, I would like to know to what degree a sentence transmits wish, hate, or fear. I looked at some studies on sentiment analysis, but haven't seen any relevant result. Most of the them are on negativity or positivity of a sentence.
Consider multi-label classification, instead of a binary sentiment your dataset would have a level/degree/probability target for each emotional label of whatever dimension size you want.
In the context of neural networks, it's just a matter of replacing
sigmoid loss layer.
If getting such large dataset with desired supervised labels is hard or not possible, one idea may be to use models trained on any sentiment datasets out there (there are even some multi-dimensional too) and apply additional un(semi)-supervised learning techniques to extract what you want.