基于相似性算法与蚁群算法的聚类算法

由于当今的网络数据是海量的,因此科研人员对某些问题进行研究时需要将不同属性的数据从中提取出来,然而在提取这些数据之前需要将相同数据进行聚类。数据聚类的过程,也就是寻找数据最优属性的过程,然而人工蚁群就是一种寻找问题最优解的算法,因此在本文中再次将蚁群算法在聚类中进行应用。由本文提出的聚类算法可以分为两个部分,第一部分是：通过相似性算法来衡量数据之间的相似度,第二部分是：根据第一部分的计算结果,再采用蚁群算法为需要聚类的数据选择不同的聚类中心,从而对不同属性的数据进行聚类,经过以上两个过程的计算,可以实现对数据的聚类。在本文中进行数据聚类时采用的相似性度量来代替距离的计算,是本文创新点之一,采用蚁群算法在聚类过程中来选择聚类中心也是本文的创新所在。

A clustering algorithm based on similarity algorithm and ant colony algorithmInformation center of Nanjing University of Aeronautics Astronautics, Jiangsu 210016, Nanjing Province

as the network data today is massive, so the researchers on the study of some problems need to be different from the properties of the data extraction, however, these data will be required clustering before extracting the same data . The process of data clustering is the process of finding the optimal attributes of data. However, the artificial ant colony is an algorithm for finding the optimal solution of the problem. Therefore, the ant colony algorithm is applied to the clustering in this paper. The clustering algorithm proposed by this paper can be divided into two parts, the first part is: through the similarity algorithm to measure the similarity between the data, the second part is: according to the results of the first part, using ant colony algorithm to select different clustering in the heart for clustering data, and clustering of the different attributes of the data. Through the calculation of the above two processes, can realize the clustering of data. In this paper, the similarity measure used in data clustering is used to replace the distance calculation. It is one of the innovations of this paper. It is also the innovation of this paper that the ant colony algorithm is used to select the cluster center in the clustering process.