基于Tabu搜索的聚类算法研究

聚类分析的两个基本任务是分析数据集中簇的数量以及这些簇的位置。大多数的聚类方法通常只关注后一个问题。为了在聚类数不确定的情况下实现聚类分析，本文提出了一种新的结合人工免疫网络和Tabu搜索的动态聚类算法—DCBIT。新算法主要包含两个阶段：先使用人工免疫网络算法获得一个候选聚类中心集，然后使用Tabu搜索在候选聚类中心集上实现动态聚类。仿真实验结果表明与现有方法相比，新方法具有更好的收敛概率和收敛速度。

A Novel Dynamic Clustering Algorithm Based on Tabu Search

Cluster analysis aims at answering two main questions: how many clusters there are in the data set and where they are located. Usually, the traditional clustering algorithms only focus on the last problem. In order to solve the two problems at the same time, this paper proposes a novel dynamic clustering algorithm called DCBIT, which is based on the immune network and Tabu search. The algorithm includes two phases, it begins by running im- mune network algorithm to find a candidate clustering center set, and then it employs Tabu search to search the opti- mum number of clusters and the location of each cluster according to the candidate centers. Experimental results show that the hew algorithm has satisfied convergent probability and convergent speed.