Show description

Analysis of the results shows a clear relation between certain aspects of the topic and task on one side and the most frequently used CAs on the other side. Some first conclusions from those experiments are the following. ) than segments about topics Meeting Modelling in the Context of Multimodal Research 29 dealing with past or present or atemporal events. Segments dealing with topics that directly relate to the speaker or the whole group as opposed to topics about something entirely external also contain significantly more action motivators.

This implies, however, that if the model changes, the annotations are influenced as well and vice versa. The choice of annotation schemas and structures for the separate boxes should in most applications be inspired by explanatory models of humans inter- Meeting Modelling in the Context of Multimodal Research 31 action and the application goals. Different models or different uses of the models may lead to distinct annotation schemas for the information in the boxes. 1 Manual Annotations The annotations discussed above are not necessarily automatically produced: corpus based work always involves a large amount of manual annotation work as well.

In its simplest form this could, for example, be the purely emotional rejection of a statement using the expression “I don’t feel it that way”, which, although it contains no logical argument, can influence the outcome of the discussion. Carofiglio and De Rosis [26] state that argumentation knowledge and emotion should be combined to select the most promising strategy to reach one’s goals. Information about previous meetings might, as background knowledge, be relevant for understanding some of the argument structures.