Expertise Location using Automatically Generated Network Models

Period of Performance: 10/12/2000 - 10/12/2001


Phase 1 STTR

Recipient Firm

154 Whitetail Dr
Ithaca, NY 14850
Principal Investigator
Firm POC

Research Institution

Cornell University
426 Phillips Hall
Ithaca, NY 14853
Institution POC


Many information search tasks involve finding a trusted human expert rather than simply a set of documents. Examples include finding experts on technical topics in a large organization such as the military or other governmental body, putting together project teams in an R&D organization, and contracting with freelance programmers for short-term assignments. In order to solve such tasks efficiently it is necessary to understand the social context of expertise in the organization: that is, knowledge of both the expertise of different individuals (e.g. who is an expert on air campaign planning) and of the relationships between individuals (e.g. who are colleagues, who has worked for whom in the past, etc.). Because this kind of highly structured information is rarely directly available expertise location is traditionally a costly manual task. We are building tools to support automated information and expertise location based on techniques from data mining, artificial intelligence, and graph analysis. Components include: (1) tools for gathering unstructured data about people, projects, and organizations from multiple sources; (2) data mining algorithms for automatically structuring this information both in terms of expertise profiles of individuals and a social network model of the relationships between individuals; and (3) search and visualization tools for efficiently finding experts based on this structured data. A major potential benefit of this technology will be to facilitate the flow of information within large, highly distributed, dynamically changing organizations. It will aid in quickly forming high-quality project teams, training new recruits and relocated personnel, and finding the best internal and external consultants, resulting in faster response times to opportunities and emergencies. In additional to the potential for directly marketing these tools to government/corporations for internal use, we are using them to develop our own commercial web sites exploiting these ideas.