China Communications • December 2014 54 ries in which the members are most similar to each other[1].Most clustering methods can be classifed into the following categories[2]: • Partitioning method. • Hierarchical method. • Density-based method. • Grid-based method. • Other clustering algorithm methods. K-means algorithm is a partitioningcluster- ing algorithm that due to its high power and simplicity is widely used in data clustering[3]. The main goal of this algorithm is that the total dissimilarity between all objects in a cluster is less than the corresponding clusters centers; In other words, the distance between each data point to its cluster center is less than the distance to any other cluster centers.This algorithm considers clustering the database to a certain number of clusters, so that the mean squared error (MSE) is minimized[4].K-means algorithm starts with determining the initial centers[5].These centers should be selected in such a way that anadequate distance from each other to create thegood results [4].So it can be said that the success of the k-means algorithm depends on correctly selecting initial cluster centers and also the number of the clusters. K-means algorithm, despite its simplicity and power, suffers from the following prob- lems[6-8]: 1- Final solution of the algorithm depends on the choice of initial centers (conver- Abstract: Clustering is one of the most widely used data mining techniques that can be used to create homogeneous clusters. K-means is one of the popular clustering algorithms that, despite its inherent simplicity, has also some major problems. One way to resolve these problems and improve the k-means algorithm is the use of evolutionary algorithms in clustering. In this study, the Imperialist Competitive Algorithm (ICA) is developed and then used in the clustering process. Clustering of IRIS, Wine and CMC datasets using developed ICA and comparing them with the results of clustering by the original ICA, GA and PSO algorithms, demonstrate the improvement of Imperialist competitive algorithm. keywords: data mining; homogeneous cluster; imperialist competitive algorithm I. INTRODUCTION In the current world, where huge volumes of data has been generated and accumulated, analyzing each data alone, is time-consuming and cumbersome.The solution of this problem is to classify data into the clusters and analyze them. To carry out this strategy, clustering techniques have been developed.Clustering is one of the most useful techniques in data mining with the purpose of classifying objects, documents and etc. to homogeneous catego- A New Method for Clustering Based on Development of Imperialist Competitive Algorithm Mohammad Reza Dehghani Zadeh, Mohammad Fathian, Mohammad Reza Gholamian School of Industrial Engineering, Iran University of Science and Technology, Narmak, 16844, Tehran, Iran BIG DATA