Igor Perisic

Is there a space for social computing within enterprise applications?

Biography
Dr. Igor Perisic heads Entopia’s research program focusing on 1) semantic content extraction of unstructured texts through advanced neural networking and vector space technology, 2) customized taxonomy generation 3) social networks modelisation and 4) data-mining techniques such as clustering and classifications. The French-speaking Swiss native, Dr. Perisic is currently designing a program to ensure state-of-the-art semantic search tools leveraging the natural social networks of corporations for Entopia products and services.

Dr. Perisic holds a bachelor of science in mathematics from the Swiss Federal Institute of Technology and a doctorate in statistics from Harvard University in Cambridge, Massachusetts. Dr. Perisic joined Entopia from the University of Connecticut’s Department of Statistics, where he was an Instructor and Assistant Professor. At that time, Dr. Perisic also served as a Statistical Consultant for the University’s Ergonomic Technology Center, Institute for Violence Reduction and Health Center divisions of Medicine and Occupational and Environmental Medicine, respectively. From 1998 to 2000, Dr. Perisic was a Statistical Consultant for Talaria, Inc., where he co-investigated the development of new techniques for exploratory data analysis. Prior to joining the University of Connecticut, Dr. Perisic was a Harvard Medical School Channing Laboratory Predoctoral Fellow.

Dr. Perisic has published several scientific articles related to the educational, Biostatistical, data-mining and information retrieval fields. Dr. Perisic has also been the recipient of numerous awards, including several NSF and NIH funds, and Swiss National Science Foundation grants.

Position Paper
Is there a space for social computing within enterprise applications? While the main thrust of social computing is around the open world, can an enterprise harvest the systematic interactions, i.e. social networks, between its employees to better its position within their respective markets?

We propose to look at the potential, partially realized but mostly forthcoming, of Social Networks Analysis for an Organization, in other words: Enterprise Social Network Analysis. A developed application is to integrate a set of Social Networks Analysis applications, using the traditional input of any search engine, and extract a picture of interconnectedness between entities or “knowledge workers”. Following this approach we are able to provide a clear insight on information flow and knowledge transfer within an organism, that is not hindered by any recollection or framing biases. Furthermore we also use these relations to render a dynamic expertise profiling solution.

While this specific tool is already useable “as is”, some research should be to unlock the full potential of the underlying paradigm. I would like to address two aspect of such a research. The first is privacy and anonymity, which are both linked to the identity of individuals that are presented and affects the social metadata that is used in these type of applications. Also, an avatar has an explicit identity, how is this really affected by privacy and anonymity concerns? The second is about how these applications should or could be used to provide hindsights to enterprises about their business. Businesses have clear priorities and constrains, such as product deliveries or retirement issues, these can be translated into objective functions to be evaluated on the graph representing the social network.

Back to Social Computing Symposium 2005