【作者】 尚海昆；

【导师】 孟建良；

【作者基本信息】 华北电力大学（河北） ， 计算机应用技术， 2009， 硕士

【摘要】 聚类分析是数据挖掘的一个重要的研究领域,是一种用于数据划分或分组的重要手段。K-means算法是一种传统的基于划分的聚类算法,其对大规模数据进行聚类时效率较高,从而被广泛应用在数据挖掘领域。本文在研究传统聚类算法的基础上,给出基于加权蚁群聚类的WAC K-means算法。该算法首先将加权思想引入到蚁群算法当中,而后将蚂蚁的转移概率引入到K-means聚类算法中,根据概率来决定数据归属,最后将改进的WAC K-means聚类算法运用到供电企业CRM系统的客户细分研究和实际应用中,实现了对用电客户群的细分,得出了有价值的信息。更多还原

【Abstract】 Clustering is an important area for research in Data Mining, which is also an important method in data partition or data grouping. K-means algorithm is a traditional partition clustering method. It is widely used in the area of Data Mining to cluster large data sets due to its high efficiency. Based on the traditional clustering algorithms, we bring forward the WAC K-means algorithm based on the Weighted Ant Clustering. In this improved method a weighting idea is introduced to the ant algorithm and then the transition probability of ants is introduced into the K-means clustering algorithm to determine which group the data belongs to. Finally, apply the improved WAC K-means algorithm into the customer segmentation and application in Power Supply Enterprise. And it has realized the application of enterprise’s customer segmentation and we could get valuable information.更多还原