A fuzzy clustering algorithm based on the k-nearest neighbors rule for the detection of evolution

Monitoring a human operator performing a task on a technological system may be an important issue. In that case, it is more important to detect evolutions of the operator status, rather than his status itself. In this paper, the authors present a fuzzy clustering algorithm based on the k-nearest neighbors decision rule in which time is taken into account. An application to the detection of a car driver´s behavior is presented