Open Access

ARTICLE

A Clustering Method Based on Brain Storm Optimization Algorithm

by Tianyu Wang, Yu Xue, Yan Zhao, Yuxiang Wang, Yan Zhang, Yuxiang He

Nanjing University of Information Science & Technology, Nanjing, 210044, China

* Corresponding Author: Yuxiang Wang. Email:

Journal of Information Hiding and Privacy Protection 2020, 2(3), 135-142. https://doi.org/10.32604/jihpp.2020.010362

Received 02 August 2020; Accepted 29 August 2020; Issue published 18 December 2020

Abstract

In the field of data mining and machine learning, clustering is a typical issue which has been widely studied by many researchers, and lots of effective algorithms have been proposed, including K-means, fuzzy c-means (FCM) and DBSCAN. However, the traditional clustering methods are easily trapped into local optimum. Thus, many evolutionary-based clustering methods have been investigated. Considering the effectiveness of brain storm optimization (BSO) in increasing the diversity while the diversity optimization is performed, in this paper, we propose a new clustering model based on BSO to use the global ability of BSO. In our experiment, we apply the novel binary model to solve the problem. During the period of processing data, BSO was mainly utilized for iteration. Also, in the process of K-means, we set the more appropriate parameters selected to match it greatly. Four datasets were used in our experiment. In our model, BSO was first introduced in solving the clustering problem. With the algorithm running on each dataset repeatedly, our experimental results have obtained good convergence and diversity. In addition, by comparing the results with other clustering models, the BSO clustering model also guarantees high accuracy. Therefore, from many aspects, the simulation results show that the model of this paper has good performance.

---

Keywords

---