摘要
针对近邻传播(AP)算法中偏向参数与收敛系数对AP算法的聚类效果的局限性的问题,提出了一种基于粒子群的近邻传播算法(PSO-AP算法).通过将AP算法中的偏向参数与收敛系数作为粒子,然后使用粒子群算法来对其进行智能地调整,进而提高AP算法的聚类效果.实验结果表明,该算法能有效地解决偏向参数与收敛系数对AP算法的聚类效果局限性,提高了聚类效果与收敛精度.
Aiming at the problem that the preference parameter and damping parameter in affinity propagation algorithm have limitations to the result of clustering, this paper puts forward an affinity propagation algorithm which based on particle swarm optimization (PSO-AP). By taking the two parameters in algorithm as a particle, then adjust it Intelligently by particle swarm optimization (PSO) algorithm, and improve the effect of clustering. The results of experiment show that the algorithm has effectively solved the problem, improved the result of clustering and the accuracy of damping.
出处
《计算机系统应用》
2014年第3期103-107,76,共6页
Computer Systems & Applications
基金
浙江省教育厅科研项目(Y201326770)
浙江省教育厅科研项目(Y201326872)
宁波大学科研基金项目(XYL12009)