Abstract
Individuals in organizations often need to access knowledge that is outside their own area of expertise. Doing so requires “know-who” knowledge–knowledge of who knows what inside (or outside) the organization. This paper develops the concept of a knowledge directory, a database that can be searched via query. The knowledge directory returns a list of individuals rank ordered by the mutual affinity between their own areas of expertise and the content of the query. The technology behind the matching of person and query is centering resonance analysis, which develops a knowledge map, representing objects embedded in text as a network graph. We also present a way to analyze a collection of knowledge maps so that clusters of people who have similar expertise may be identified. We apply these techniques to a group of ten university faculty from the areas of industrial engineering, operations management, and supply chain management. Experiments attempting to match queries to specific faculty members are presented and discussed, and clustering techniques are used to show how the faculty aggregate into cliques of similar content and methodological expertise. Extensions of the method are discussed.