Recognizing Emotions in Short Texts

Emotion detection is increasingly used in Embodied Conversational Agents to create an adapted reply channel to the user's affective state. In this context, we propose a method to detect emotions in short texts (i.e. in texts whose size is similar to dialog utterances). Our goal is to design a model to detect the dominant affective state produced by short texts onto a reader and to classify them into six clusters, corresponding to Ekman's psychological theory. In the current paper, the corpus consists of newspaper headlines, from SemEval 2007, task 14 [SM08] The corpus was chosen because of the appropriate size of its elements and their high emotional content. Since the methods presented in the paper, related to the corpus do not offer a good accuracy, we introduce a new classification mechanism based on the Self Organizing Maps. Also, our approach can be easily transposed to other contexts such as chat logs, forums or oral transcripts.