Exploring the Compositionality of Emotions in Text: Word Emotions, Sentence Emotions and Automated Tagging

This paper presents an approach to automated marking up of texts with emotional labels. The approach considers the representation of emotions as emotional dimensions. A corpus of example texts previously annotated by human evaluators is mined for an initial assignment of emotional features to words. This results in a List of Emotional Words (LEW) which becomes a useful resource for later automated mark up. An algorithm for the automated mark up of text is proposed. This algorithm mirrors closely the steps taken during feature extraction, employing for the actual assignment of emotional features a combination of the LEW resource, the ANEW word list, and WordNet for knowledge-based expansion of words not occurring in either. The algorithm for automated mark up is tested against texts from the original samples used for feature extraction to test its correctness. It is also tested against new text samples to test its coverage. The results are discussed with respect to two main issues: correctness and coverage of the proposed algorithm, and additional techniques and solutions that may be employed to improve the results.