查询结果:   宁念文,许合利,刘喜峰.基于资源分配指标的最大约束社区发现算法[J].计算机应用与软件,2017,34(7):217 - 221,297.
中文标题
基于资源分配指标的最大约束社区发现算法
发表栏目
算法
摘要点击数
410
英文标题
MAXIMUM CONSTRAINED COMMUNITY DETECTION ALGORITHM BASED ON RESOURCE ALLOCATION INDEX
作 者
宁念文 许合利 刘喜峰 Ning Nianwen Xu Heli Liu Xifeng
作者单位
河南理工大学计算机科学与技术学院 河南 焦作 454000 郑州工程技术学院机电与车辆工程学院 河南 郑州 450044    
英文单位
College of Computer Science and Technology, Henan Polytechnic University, Jiaozuo 454000,Henan,China Institute of Electrical and Mechanical Vehicle Engineering, Zhengzhou Institute of Technology, Zhengzhou 450044,Henan,China    
关键词
社区发现 模块度最大化 资源分配指标 最大约束标记传播模型
Keywords
Community detection Modularity optimization Resource allocation Maximum constraint label propagation model
基金项目
国家自然科学基金项目(61202286);国家科技重大专项核心电子器件、高端通用芯片及基础软件产品专项(2014ZX01045-102)
作者资料
宁念文,硕士生,主研领域:数据挖掘,并行计算。许合利,教授。刘喜峰,教授。 。
文章摘要
在复杂网络中的社区发现一直受到广泛的关注,基于模块度最大化的方法是目前流行的社区发现技术。提出一种基于资源分配(RA)指标和多步贪婪凝聚策略的模块度最大化社区发现算法RALPA(Resource Allocation-based of Label propagation Algorithm)。该算法利用准确衡量节点间相似性的RA指标,通过最大约束标记传播模型使社区内部节点拥有较高的相似性,与社区外部的节点拥有较低的相似性。然后,通过多步贪婪凝聚策略将划分模块度增加最大的多对小社区进行合并。实验结果表明,该算法不仅避免了对节点更新顺序的敏感和易得到平凡解的问题,而且提高了算法的稳定性和社区划分的精度。
Abstract
Community detection in complex networks have received wide attention, and the method based on modularity maximization is the popular community detection technology. In this paper, a modularity maximization community detection algorithm named RALPA (Resource Allocation-based of Label Propagation Algorithm) is proposed, which is based on resource allocation (RA) and multi-step greedy cohesion strategy. The algorithm uses the RA index to measure the similarity between nodes accurately. By using the maximum constraint label propagation model, the internal nodes of the community have high similarity, and have low similarity with the nodes outside the community. Then, through the multi-step greedy cohesion strategy, the multi-pair small communities with the largest increase of partitioning degree will be merged. The experimental results show that the proposed algorithm not only avoids the problem of the sensitivity of node update order and the trivial solution, but also improves the stability of the algorithm and the accuracy of community division.
下载PDF全文