Experience-Based Resource Description and Selection in Multiagent Information Retrieval

In this paper, we propose an agent-centric approach to resource description and selection in a multiagent information retrieval (IR). In the multiagent system, each agent learns from its experience through its interactions with other agents their capabilities and qualifications. Based on a distributed ontology learning framework, our methodology allows an agent to profile other agents in a dynamic translation table and a neighborhood profile, which together help determine resource description and selection process. Further, we report on the experiments and results of the first phase of our research, which focuses on the operational issues (e.g., real-time constraints, frequency of queries, number of threads, narrowness in ontology) on how the agents handle queries collaboratively.

• Type: Conference
• Year: 2004
• Where: FLAIRS
• Authors: Leen-Kiat Soh
• Keywords: Artificial Intelligence, FLAIRS 2004, Multiagent Information Retrieval, Ontology Learning Framework, Resource Description