Ramesh Vasudevan

Director, Software Development

Oracle Corporation

Redwood Shores, California, USA

email: ramesh.vasudevan-at-oracle.com

I manage a team that works on a few aspects of the Oracle Beehive Collaboration Server product. Over the past year or so, my team has been working on adding Expertise Management and Social Computing features to the Beehive product.

The Expertise Management server crawls artifacts (such as Email messages, Documents, Wiki pages, Blogs, etc) authored by users, converts the binary content in various formats (such as word, pdf, etc) to text as needed, performs linguistic analysis of the content to extract noun phrases, and builds a profile of the user that captures the user's interests and expertise with a confidence attached to each phrase. Users can subsequently search for experts in specific areas and the server acts as a broker to route these search requests in a single-blind fashion to matched user profiles i.e. the matched user's identity is not revealed to the searcher unless the user accepts the request. This unique brokering model ensures privacy while allowing deployments to tap into the knowledge dispersed across the organization.

The Social Computing framework we added to Beehive includes support for social networks: explicit (mutual connections as well as non-mutual following), implicit (such as the reporting structure), and derived networks (relationships derived by analyzing artifacts such as email messages, calendar meetings, etc). The framework leverages social networks to provide activity feeds, social bookmarking, popular content (most active, most bookmarked, etc from your connections), recommended content & people (combine my activity with my network's activity to recommend content and people), etc

source: http://www.linkedin.com/in/rameshvasudevan