By analyzing explicit & implicit feedback information retrieval systems can determine topical relevance and tailor the search criteria to the user needs. In this talk, I will explore whether it is possible to infer what is relevant for
a user in an information retrieval task, through the observation
of their aaffective behavior. The sensory data employed 
in our study vary between facial expressions and other peripheral
physiological signals, which are all regarded as indicative
of the user's current aaffective state. The modeling
goal is to predict with reasonable accuracy topical relevance.
We extract a set of features from the signals and explore the
data using classification methods, such as SVM and KNN.
The results of our initial evaluation indicate that prediction
of relevance is possible, to a certain extent, and models can
benefitt from taking into account user affective behavior.