Title :
Mining Association Rules in Scale-Free Networks
Author :
Gao, Li ; Dai, Shang-Ping ; Zhu, Chang-Wu
Author_Institution :
Hua Zhong Normal Univ., Wuhan
Abstract :
For the characteristic of scale-free networks, containing a few nodes that have a very high degree and many with low degree,the high connectivity nodes play an important role of hubs in communication and networking. This characteristic can be exploited with designing efficient search algorithms. This paper proposes an algorithm to change each new node connecting to the network based on its high-degree-probability for equal-degree-probability ,after having constituted initial model by choosing high-degree-probability nodes. We use an association rule search strategy that utilizes high degree nodes in scale-free networks and costs scaling with the size of the graph. We also demonstrate the utility of these CSCCNU network.. It can improve networks´ robustness.
Keywords :
data mining; probability; search problems; association rule search strategy; equal-degree-probability; high-degree-probability; mining association rules; scale-free networks; search algorithms; Algorithm design and analysis; Association rules; Computer science; Cybernetics; Data mining; Electronic mail; IP networks; Intelligent networks; Joining processes; Machine learning; Association rules; Data mining; Genetic algorithm; Hubs; Scale-free networks;
Conference_Titel :
Machine Learning and Cybernetics, 2007 International Conference on
Print_ISBN :
978-1-4244-0973-0
Electronic_ISBN :
978-1-4244-0973-0
DOI :
10.1109/ICMLC.2007.4370248