一种高效的基于教与学的社区发现算法

(厦门大学信息科学与技术学院,福建 厦门 361005)

社区发现; 教与学; 多目标; 多种群进化算法

An Efficient Multi-population Community Detection Algorithm Using Teaching-learning-based Optimization
LI Peixi,FENG Shaorong*

(School of Information Science and Engineering,Xiamen University,Xiamen 361005,China)

community detection; teaching-learning; multi-objective; multi-population evolutionary algorithm

DOI: 10.6043/j.issn.0438-0479.201709025

备注

社区结构是复杂网络的重要特征,社区发现就是为了挖掘复杂网络中的社区结构.为了提高基于教与学的多目标社区发现算法(MODTLBO/D)的准确率,降低时间复杂度,提出了一种在多种群进化策略下的MODTLBO/D(E-MODTLBO/D).在E-MODTLBO/D中,采用自适应学习因子加强在教学阶段的探索与搜索能力; 在学习阶段,每个个体在各自的子种群内采用随机学习策略或者是改进的量子行为学习策略.在每次迭代更新后,子种群间进行信息交流,维持算法的多样性与避免早熟收敛.实验表明,E-MODTLBO/D在时间复杂度与发现高质量的社区结构方面要优于MODTLBO/D等一些经典社区发现算法.

Community structure is an important feature of complex networks,and community detection aims at mining the community structure of complex networks.In order to improve the multi-objective optimization of community detection using discrete tea-ching-learning-based optimization with decomposition(MODTLBO/D),and decrease time complexity,we propose an efficient tea-ching-learning-based optimization algorithm combined with multi-population evolutionary strategy for community detection.In this study,we adopt adaptive learning factor in teacher phase to enhance the ability of exploration and search.In learner phase,each learner employs the random learning strategy or modified quantum-behaved learning strategy in corresponding subpopulation.After each generation,subpopulations exchange information to maintain the diversity and discourage premature convergence.The experiments results demonstrate that our proposed algorithm has an advantage of time complexity and is highly efficient at discovering quality community structure.