-->

KMWorld 2024 Is Nov. 18-21 in Washington, DC. Register now for Super Early Bird Savings!

  • October 17, 2001
  • News

Seek and ye shall find experts

In a move designed to allow users to personalize their searches and automatically identify experts within their organization, Applied Psychological Research (APR) has formed an alliance with Convera that will bring together what they call intelligent user profiling with knowledge retrieval. The solution will combine APR’s Youmeus software with Convera’s RetrievalWare.

Convera explains RetrievalWare is a cross-lingual, multimedia solution that uses pattern recognition and semantic network technologies to examine and understand the content of data held in 250 different file formats. Convera says the semantic network contains approximately 500,000 word meanings and 1.6 million word links, and it is able to recognize spelling mistakes in searches, focusing on word meaning and relationships to locate suitable documents.

APR says Youmeus generates profiles of users based on their explicit requirements, as well as on their implicit tastes and preferences expressed through their behavior when interacting with systems and touchpoints. The company adds that Youmeus identifies users’ needs in terms of consistent personality characteristics as well as current, transient requirements. APR claims profiles generated ensure that both types of needs are understood and that only the most relevant information is routed to users. Profiles are also used to identify patterns and clusters of users with similar needs, tastes, preferences and behavior. The information recommended by the software is based on a match with the user's search criteria, their personal profile and on what similar users have found interesting.

Youmeus further analyses data flow between individuals to provide an overview of each employee's information usage. Experts in particular subject areas are automatically identified and queries from other users are routed to them.

KMWorld Covers
Free
for qualified subscribers
Subscribe Now Current Issue Past Issues