一种基于节点影响力的局部社区发现算法

发布时间：2018-05-07 10:24

本文选题：社区结构 + 点权　；参考：《小型微型计算机系统》2013年09期

【摘要】：为快速准确寻找社会网络中的社区结构,从节点影响力的角度出发,提出一种新的社区发现算法.算法设计过程中引入了点权,它能够衡量节点影响力的大小;首先根据"种子"节点的点权有选择地进行广度优先搜索,使点权较大的节点不断地影响点权较小的节点,进而得到"种子"节点所在的社区结构,然后再从已知的社区外任取一个新的"种子"节点,重复上述过程,就可得到整个网络的社区结构.对算法进行优化并应用到实际网络,实验结果验证了算法的可行性,与经典算法相比,该算法的准确性和计算速度都有所提高.
[Abstract]:In order to find community structure in social network quickly and accurately, a new community discovery algorithm is proposed from the point of view of node influence. In the process of algorithm design, the point weight is introduced, which can measure the influence of the node. Firstly, according to the point weight of the "seed" node, the breadth-first search is carried out selectively, so that the node with the larger point weight constantly affects the node with the smaller point weight. Then the community structure of the "seed" node is obtained, and then a new "seed" node is taken from the known community, and the community structure of the whole network can be obtained by repeating the above process. The algorithm is optimized and applied to the actual network. The experimental results verify the feasibility of the algorithm. Compared with the classical algorithm, the accuracy and computing speed of the algorithm are improved.
【作者单位】：东北大学信息科学与工程学院;
【基金】：国家自然科学基金项目(61070162,60903159)资助国家高新技术研究发展计划重点项目(2007AA041201)资助国家科技支撑计划项目(2008BAH37B05)资助中央高校基本科研业务费项目(N110216001)资助
【分类号】：TP393.0

【参考文献】