Exploiting Enterprise Models for the Automatic Distribution of Corporate Information

Abstract

Effectively distributing information to members of a large organization poses a number of challenges. Ideal information distribution within an organization will be timely, selective, and (to some degree) automatic. It is difficult to meet all three of these criteria simultaneously using existing techniques in information retrieval and information filtering. This paper introduces a new method of distributing information automatically, based on enterprise models—representations of the pertinent aspects of an organization’s structure and operation. The approach is based on the premise that often information needs within an organization are implicitly represented by paths through the enterprise model. These paths are discovered in our approach through the use of Generalized Path Models (GPMs). A GPM specifies (1) the conditions that instigate a path search when new information is created, (2) a general class of paths for which to search, and (3) an action to take when a suitable path is found. The approach has been implemented in a software module called IDIST, an extension to a large commercial enterprise modelling tool. Keywords: Artificial Intelligence, Artificial Intelligence Center, AIC