Fuzzy Miner: Extracting Fuzzy Rules from Numerical Patterns
Nikos Pelekis (Greece & UMIST Manchester, UK), Babis Theodoulidis (UMIST Manchester, UK), Ioannis Kopanakis (UMIST Manchester, UK) and Yannis Theodoridis (University of Piraeus, Greece)
Copyright: © 2008
QOSP Quality of Service Open Shortest Path First based on QoS routing has been recognized as a missing piece in the evolution of QoS-based services in the Internet. Data mining has emerged as a tool for data analysis, discovery of new information, and autonomous decision-making. This paper focuses on routing algorithms and their appli-cations for computing QoS routes in OSPF protocol. The proposed approach is based on a data mining approach using rough set theory, for which the attribute-value system about links of networks is created from network topology. Rough set theory offers a knowledge discovery approach to extracting routing-decisions from attribute set. The extracted rules can then be used to select significant routing-attributes and make routing-selections in routers. A case study is conducted to demonstrate that rough set theory is effective in finding the most significant attribute set. It is shown that the algorithm based on data mining and rough set offers a promising approach to the attribute-selection prob-lem in internet routing.