Improving agent-based meeting scheduling through preference learning.

This paper presented an autonomous Secretary Agent (SA) that can perform meeting scheduling task on behalf of their respective user through negotiations. Previous study of searching strategy uses relaxation process to allow agents negotiate by relaxing their preference when conflict arises. However, this increased the cost of searching process. As a result, an improvement of relaxation searching strategy by adapting Neural Network (NN) learning mechanism is proposed. The back-propagation learning method is used in this research to intelligently predict the participants’ preferences and guide the host in selecting proposal s that are more likely to get accepted. Hence, higher quality solution can be found in lower communication cost. The comparison result between the proposed and two previous estimation strategies showed improvement of quality of the solution as well as the communication cost of the proposed strategy.

Actions (login required)

View Item