Deep learning for affective computing: text-based emotion recognition in decision support
This work addresses the problem of enhancing emotion recognition in text for affective computing applications, such as decision support systems, but it is incremental as it builds on existing deep learning methods with specific customizations.
The paper tackled the challenge of accurately recognizing emotions in narrative documents for decision support by customizing recurrent neural networks with bidirectional processing, dropout regularization, and weighted loss functions, and introducing sent2affect, a transfer learning approach from sentiment analysis to emotion recognition, achieving consistent performance improvements over traditional machine learning across 6 benchmark datasets.
Emotions widely affect human decision-making. This fact is taken into account by affective computing with the goal of tailoring decision support to the emotional states of individuals. However, the accurate recognition of emotions within narrative documents presents a challenging undertaking due to the complexity and ambiguity of language. Performance improvements can be achieved through deep learning; yet, as demonstrated in this paper, the specific nature of this task requires the customization of recurrent neural networks with regard to bidirectional processing, dropout layers as a means of regularization, and weighted loss functions. In addition, we propose sent2affect, a tailored form of transfer learning for affective computing: here the network is pre-trained for a different task (i.e. sentiment analysis), while the output layer is subsequently tuned to the task of emotion recognition. The resulting performance is evaluated in a holistic setting across 6 benchmark datasets, where we find that both recurrent neural networks and transfer learning consistently outperform traditional machine learning. Altogether, the findings have considerable implications for the use of affective computing.