Contracting community for computing maximum flow

• Published in: 2012 IEEE International Conference on Granular Computing pp. 651 - 656
• Main Authors: Yanping Zhang, Xiansheng Xu, Bo Hua, Shu Zhao
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.08.2012
• Subjects: Artificial neural networks; community; contracting; Educational institutions; maximum flow; network
• DOI: 10.1109/GrC.2012.6468649
• ISBN: 1467323101; 9781467323109

In this paper, we propose a novel method named Contracting Community Approach (CCA) to get the maximum flow of flow network. Firstly, we contract communities in the original network. Then, we apply classic algorithms on the contracted network to approximately solve the maximum flow problem. Experimental results show that the efficiency of the proposed algorithm. For sparse networks, the size of network is reduced to 58.38% averagely and the correctness of maximum flow is over 95%. For middle dense networks, the size of network is reduced to 65.77% averagely. For dense networks, the size of network is reduced to 64.84% averagely. And the correctness of maximum flow even reach 100% both in many middle dense and dense cases in our experiments.