Abstract. It is difficult to track, parse and model human-computer interactions during editing and revising of documents, but it is necessary if we are to develop automated technologies that will aid or replace humans. This paper introduces a system for accessing and recording a stream of events related to human actions in a real-time cartographic map revision system. The recorded events are parsed into a sequence of meaningful user actions and an action representation in XML format is generated. We also report results of experiments on predicting user actions such as view changes, edits, road tracking/production using hidden Markov models.
Jun Zhou, Walter F. Bischof, Terry Caelli