ABSTRACT
People can read a line or two of a story and immediately know “what is going on.” So in reading
As the boy walked down the aisle he took a can of tuna fish from the shelf and put it in his basket.
we have no trouble deciding that we are witnessing a person shopping at a supermarket. When I talk of “context recognition” this is what I mean. The question I wish to raise is: How do we do this?
We will attack this question in a somewhat oblique fashion. After outlining a knowledge representation designed to facilitate certain kinds of inference, we will show how it can handle the more familiar “script application” situations, in which we recognize why a person performed an action (e.g. light a cigarette) because a certain “frame” or “script” is active (e.g. “smoke cigarette”). We will then show how a trivial modification enables us to handle many context recognition problems.
