当前位置:
X-MOL 学术
›
J. Comb. Optim.
›
论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
KATZ centrality with biogeography-based optimization for influence maximization problem
Journal of Combinatorial Optimization ( IF 1 ) Pub Date : 2020-05-08 , DOI: 10.1007/s10878-020-00580-6 Abbas Salehi , Behrooz Masoumi
Journal of Combinatorial Optimization ( IF 1 ) Pub Date : 2020-05-08 , DOI: 10.1007/s10878-020-00580-6 Abbas Salehi , Behrooz Masoumi
In the field of social networks, the Influence Maximization Problem (IMP) is one of the most well-known issues that have attracted many researchers in recent years. Influence Maximization (IM) means trying to find the best subset of K nodes that maximizes the number of nodes influenced by this subset. The IM is an NP-hard problem that plays an important role in viral marketing and dissemination of information. The existing solutions like greedy approaches to solving IMP do not have the efficiency and accuracy in solving the problem. In this paper, we propose a new metaheuristic algorithm based on Katz centrality with biogeography-based optimization to solve IMP in the social network. In the proposed algorithm, each habitat with the subset of K nodes is considered as the solution to the IM problem. In the proposed algorithm, the Katz centrality of each node is calculated and used as the emigration rate of each habitat. The focus of the study has been on improving the performance of the BBO algorithm by combining it with the Katz centrality. The objective was to use an enhanced meta-heuristic algorithm with measuring centrality to solve the IM problem. In the results of experiments based on different types of real-world social networks, it is well known that the proposed algorithm is more efficient, accurate, and faster than influence maximization greedy approaches.
中文翻译:
KATZ中心结合基于生物地理的优化来解决影响最大化问题
在社交网络领域,影响最大化问题(IMP)是近年来引起众多研究人员关注的最著名问题之一。影响力最大化(IM)意味着试图找到K个节点的最佳子集,该子集可以最大化受此子集影响的节点数。IM是一个NP难题,在病毒式营销和信息传播中起着重要作用。现有的解决方案(如贪婪的解决IMP的方法)不具有解决问题的效率和准确性。在本文中,我们提出了一种基于Katz中心性和基于生物地理学的优化的新的启发式算法,以解决社交网络中的IMP。在提出的算法中,每个栖息地都带有K的子集节点被认为是IM问题的解决方案。在提出的算法中,计算每个节点的Katz中心度,并将其用作每个栖息地的迁移率。该研究的重点一直在于通过将其与Katz中心性相结合来提高BBO算法的性能。目的是使用具有中心度的增强型元启发式算法来解决IM问题。在基于不同类型的现实世界社交网络的实验结果中,众所周知,与影响最大化贪婪方法相比,该算法更有效,更准确,更快速。
更新日期:2020-05-08
中文翻译:
KATZ中心结合基于生物地理的优化来解决影响最大化问题
在社交网络领域,影响最大化问题(IMP)是近年来引起众多研究人员关注的最著名问题之一。影响力最大化(IM)意味着试图找到K个节点的最佳子集,该子集可以最大化受此子集影响的节点数。IM是一个NP难题,在病毒式营销和信息传播中起着重要作用。现有的解决方案(如贪婪的解决IMP的方法)不具有解决问题的效率和准确性。在本文中,我们提出了一种基于Katz中心性和基于生物地理学的优化的新的启发式算法,以解决社交网络中的IMP。在提出的算法中,每个栖息地都带有K的子集节点被认为是IM问题的解决方案。在提出的算法中,计算每个节点的Katz中心度,并将其用作每个栖息地的迁移率。该研究的重点一直在于通过将其与Katz中心性相结合来提高BBO算法的性能。目的是使用具有中心度的增强型元启发式算法来解决IM问题。在基于不同类型的现实世界社交网络的实验结果中,众所周知,与影响最大化贪婪方法相比,该算法更有效,更准确,更快速。