Affect Detection from Social Contexts Using Commonsense Knowledge Representations

In the past years, an important volume of research in Natural Language Processing has concentrated on the development of automatic systems to deal with affect in text. In spite of this interest, the performance of the approaches is still very low. An explanation to this fact is that emotion is most of the times not expressed through specific words, but by evoking situations that have an affective meaning. Dealing with this phenomenon requires automatic systems to have "knowledge"on the situation, the concepts it describes and their interaction. This necessity motivated us to develop the EmotiNet knowledgebase - a resource for the detection of emotion from text based on commonsense knowledge on concepts, their interaction and their affective consequence. In this article, we present an overview of the process undergone to build EmotiNet, propose methods to extend the knowledge it contains and analyze the performance of implicit affect detection using this resource. Additionally, we compare the results obtained with EmotiNet to the use of well-established methods for affect detection. The results of our extensive evaluations show that the approach using EmotiNet is appropriate for capturing and storing the structure of implicitly expressed affect, that the knowledge it contains can be easily extended to improve the results of this task and that methods employing EmotiNet obtain better results than existing methods for emotion detection.