Digitalizing the Cognitive-emotional Fugue: How to Measure Affect in Ongoing e-Negotiations?

Affective communication and emotions are an important part of negotiations. Negotiation support and negotiation support systems, however, tend to neglect this aspect given measurement difficulties. In the presentation I will explore possibilities of "measuring" affect in communication and firstly show results of state of the art supervised machine learning techniques that could enable support systems to classify emotions expressed in ongoing negotiation communication. These findings show promising performance of some preprocessing and machine learning techniques, especially in the category of activating negative emotions, which is highly important in negotiations as it might reduce the prospects of reaching an agreement.