Human-agent service matching using natural language queries: system test and training

Journal article (2006)

Authors

W. Pasman Computer Graphics and Visualisation - EEMCS
J Lindenberg Interactive Intelligence - EEMCS
Research Group
Computer Graphics and Visualisation (EEMCS) (TU Delft)
More Info
expand_more

Published Date

2006

Language

Undefined/Unknown

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Faculty

Electrical Engineering, Mathematics and Computer Science

Department

Intelligent Systems

Research Group

Computer Graphics and Visualisation

Abstract

Abstract Smart environments, ambient intelligence and intelligent agents leave the user lost between large amounts of services. Ad-hoc networks, mobile agents and mobile devices make the set of available services dynamic over time and space, increasing the user¿s problems to find the service he needs. Earlier, we presented a ServiceMatcher that can find the agent best fitting to the user¿s natural language request. This paper presents performance results of the ServiceMatcher. The test queries come from human users in a realistic scenario (see our other paper in this issue). With a short training of the agent vocabularies, over 80% correct service matches are found.