Leveraging Recursive Processing for Neural-Symbolic Affect-Target Associations
This addresses the challenge of emotional interaction in social companion robots, but it is incremental as it builds on existing neural-symbolic and recursive processing techniques.
The paper tackled the problem of explaining deep learning decisions based on affect for social companion robots by developing a hybrid neural-symbolic system that associates extracted targets with affective labels in natural language. The result was higher accuracy and interpretability compared to other methods in an aspect-based sentiment analysis task.
Explaining the outcome of deep learning decisions based on affect is challenging but necessary if we expect social companion robots to interact with users on an emotional level. In this paper, we present a commonsense approach that utilizes an interpretable hybrid neural-symbolic system to associate extracted targets, noun chunks determined to be associated with the expressed emotion, with affective labels from a natural language expression. We leverage a pre-trained neural network that is well adapted to tree and sub-tree processing, the Dependency Tree-LSTM, to learn the affect labels of dynamic targets, determined through symbolic rules, in natural language. We find that making use of the unique properties of the recursive network provides higher accuracy and interpretability when compared to other unstructured and sequential methods for determining target-affect associations in an aspect-based sentiment analysis task.