A novel method for big data analysis

Given the impact of big data on the way organizations and individuals work, the organizations need to employ new architecture, new tools and activities to cope with incoming challenges. In doing so, plenty of efforts are being done to take advantage of the large-scale frameworks. In our proposed method, the data with the highest similarity will be classified in a single category and the data points will be grouped in a batch, consequently, the data included in a group will have the highest convergence. In this approach, optimization of Imperialist Competitive Algorithm and k-means Clustering Algorithm are hired for analysis of big data. Accordingly, the former algorithm is used for optimization of the results of latter algorithm, gaining the highly-accurate classification results. Finally, these results will be compared only to those of k-means Clustering Algorithm.

Keywords:

Big data , Imperialist Competitive Algorithm , k-means Clustering Algorithm