Learning Word Ratings for Empathy and Distress from Document-Level User Responses

TL;DR

This paper introduces a novel deep learning method to automatically generate empathy lexicons from document-level ratings, providing interpretable emotion analysis tools for NLP applications.

Contribution

It develops the first deep learning approach for learning empathy word ratings from higher-level supervision and creates the first empathy lexicon for NLP.

Findings

MLFFN outperforms other methods in learning word ratings

The resulting empathy lexicon is publicly available

Insights into word groupings obtained via Signed Spectral Clustering

Abstract

Despite the excellent performance of black box approaches to modeling sentiment and emotion, lexica (sets of informative words and associated weights) that characterize different emotions are indispensable to the NLP community because they allow for interpretable and robust predictions. Emotion analysis of text is increasing in popularity in NLP; however, manually creating lexica for psychological constructs such as empathy has proven difficult. This paper automatically creates empathy word ratings from document-level ratings. The underlying problem of learning word ratings from higher-level supervision has to date only been addressed in an ad hoc fashion and has not used deep learning methods. We systematically compare a number of approaches to learning word ratings from higher-level supervision against a Mixed-Level Feed Forward Network (MLFFN), which we find performs best, and use the MLFFN to create the first-ever empathy lexicon. We then use Signed Spectral Clustering to gain insights into the resulting words. The empathy and distress lexica are publicly available at: http://www.wwbp.org/lexica.html.