Prior versus Contextual Emotion of a Word in a Sentence

A set of words labelled with their prior emotion is an obvious place to start on the automatic discovery of the emotion of a sentence, but it is clear that context must also be considered. No simple function of the labels on the individual words may capture the overall emotion of the sentence; words are interrelated and they mutually influence their affect-related interpretation. We present a method which enables us to take the contextual emotion of a word and the syntactic structure of the sentence into account to classify sentences by emotion classes. We show that this promising method outperforms both a method based on a Bag-of-Words representation and a system based only on the prior emotions of words. The goal of this work is to distinguish automatically between prior and contextual emotion, with a focus on exploring features important for this task.