摘要
聚类在数据挖掘、统计学、机器学习等很多领域都有很大应用。聚类问题可以归结为一个优化问题。人工鱼群算法(AFSA)是一种新提出的新型仿生优化算法。在分析AFSA存在不足的基础上,提出一种改进人工鱼群算法,并应用于求解聚类问题。算法保持了AFSA算法简单、易实现的特点,通过改进个体鱼的行为,并引入均匀交叉算子,将人工鱼群算法和遗传算法融合,显著提高了算法运行效率和求解质量。仿真实验取得了较好的结果。
Clustering has its roots in many areas, including data mining, statistics, and machine learning and can be regarded as an optimization problem. Artificial fish swarm algorithm (AFSA) is a novel bio inspired optimizing method. After analyzing the disadvantages of AFSA, presents an improved artificial fish swarm optimization algorithm of solving clustering analysis problem. By improving: the artificial fish's behaviors and combining artificial fish- swarm algorithm with genetic algorithm, the algorithm is as simple for implement as AFSA,but it greatly improves the ability of seeking the global excellent result and convergence property and accuracy. The simulation results show that the algorithm is more efficient.
出处
《计算机技术与发展》
2010年第3期84-87,91,共5页
Computer Technology and Development
基金
安徽省自然科学基金项目(KJ2008B021)
关键词
聚类
人工鱼群算法
交叉算子
优化
clustering
artificial fish swarm algorithm
crossover operator
optimization