Implementing an agent-based system for distributed information retrieval

With the increasingly huge amount of data located in various databases and a need for users to access them, Distributed Information Retrieval (DIR) has been at the core of preoccupations of a number of researchers. Indeed, numerous Distributed Information Retrieval systems and architectures have been proposed including the Broker-Based Architecture. Moreover, providing DIR with more flexibility and adaptability has led researchers thinking to build DIR with intelligent agents. Thus, this research proposes a design and an implementation of a novel architecture based on Broker-Based Architecture and Peer-to-Peer network called Broker-Based Peer-to-Peer Network. The proposed architecture is implemented with a Multi-agent System where a main agent playing the role of broker, receives query from a Peer Agent and forwards them to other Peer Agents each with their index and resources. The latter agents upon completing retrieval directly send their results to the Peer Agent that originated the query-instead of the Broker Agent- that collects retrieved results and finally merges them. JADE (Java Agent DEvelopment Framework) is used as a wrapper for agents and, for experiments, TERRIER (TERabyte RetRIEveR) is extended and used as search engine to retrieve TREC (Text Retrieval Conference) collections datasets notably TREC-6. The Peer Agent that originated the query progressively collects results coming from other Peer Agents, normalizes and merges them and then proceeds with re-ranking. For normalization, MinMax and Sum that are unsupervised normalization methods are used. Keywords: Distributed Information Retrieval, Multi-agent Systems, score normalization, Broker-based Peer-2-Peer Network.