Abstract
Improving searching performance is an important issue in peer-to-peer network systems. Although Distributed Hashing Tables (DHTs) can route queries more efficiently than flooding, they are not suitable to support keyword search. Emerging large-scale P2P systems employ clustering to reduce message overhead and provide system scalability. In this paper, we propose a peer-to-peer network system based on interest groups to improve search performance. Query messages are first sent to interested peers that may successfully hit the queries with high probability. Simulation results show that the proposed architecture outperforms other works in terms of message overhead, search latency, and hit ratio.