抽象的

A Decentralized Service Discovery Approach on Peer-To-Peer Networks

P.Thangalakshmi, S. Kamalesh

This project is Service-Oriented Computing (SOC) is rising as a paradigm for developing distributed applications. An essential issue of utilizing SOC is to own an ascendible, reliable, and strong service discovery mechanism. However, ancient service discovery strategies mistreatment centralized registries will simply suffer from issues like performance bottleneck and vulnerability to failures in massive ascendible service networks, therefore functioning abnormally. During this paper we have a tendency to propose a P2P call tree induction algorithmic program during which each tree peer learns and maintains the proper call compared to a centralized state of affairs. Our algorithmic program is totally decentralized, asynchronous, and adapts swimmingly to changes within the information and also the network. In meta-algorithm that uses completely different classification models in line with a learning technique known as boosting. The algorithmic program could be a learning methodology which will handle weighted instances. Finally, we propose a rule induction algorithm to produces better accuracy than the original decision tree classifier.