摘要
聚类是数据挖掘的主要任务之一,它在知识发现、模式识别、决策支持等方面有着重要应用,聚类挖掘已成为一个非常活跃的研究课题;近年来,基于智能计算的数据挖掘方法研究有了较大进展,机器学习、遗传算法、粒子群优化技术的应用在一定程度上改善和提高了聚类挖掘的性能和效率,但聚类技术仍面临着输入参数对领域知识的依赖性、交互动态性等方面的严峻挑战。
Clustering is one of most heated research topic of important DM （data mining） tasks of the day. It has many application areas such as discovery knowledge, pattern recognition, decision support system （DSS） and et al. With the rapid development of DM techniques based on intelligence computing, the application, such as machine learning, genetic algorithm, particle swarm optimization algorithm has improved and enhanced the performances and efficiency of clustering techniques. However, existing algorithms are still sensitive to data order. High effective, self- adaptive, interactively dynamic, capability for high dimension, incremental clustering algorithm should be studied. Clustering technique in data mining will yet be faced with many problems and challenges.
出处
《计算机测量与控制》
CSCD
2006年第5期561-563,582,共4页
Computer Measurement &Control
基金
安徽省高校自然科学基金资助项目(2005KJ095)
关键词
聚类
数据挖掘
智能计算
粒子群优化
遗传算法
clustering
data mining
intelligence computing
particle swarm optimization
genetic algorithm
