EmoTag is an approach to automated marking up of texts with emotional labels. The approach considers in parallel two possible representations of emotions: as emotional categories and emotional dimensions. For each representation, 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. The proposed algorithm for automated mark up of text 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, WordNet for knowledge based expansion of words not occurring in either and an ontology of emotional categories.
Corpus of tales marked up with emotions: We have selected eight tales, every one of them popular tales with different lengths (altogether they result in 10.331 words and 1.084 sentences), in English. The corpus is marked up by human evaluators with emotional dimensions and emotional categories. Download
LEW List: LEW is a list of words with its associated emotions. We have two lists, one with emotional categories and the other with emotional dimensions.
- Emotional dimensions LEW List: For emotional dimensions the LEW list stores for each pair (word’s stem, label) a value for activation, evaluation an power. This way the LEW list stores for each pair the average value of activation, evaluation and power of that pair in all the analyzed texts. Download
- Emotional categories LEW list: For emotional categories the LEW list stores for each pair (word’s stem, emotional category) the collocation factor of this pair. Download
Ontology of emotional categories: We have developed an ontology of emotional categories. They are structured in a taxonomy that covers from basic emotions to the most specific emotional categories. Download
You can find more information about EmoTag related publications in Publications.