A novel proposed deer algorithm

摘要

Deer is a social animal though it is wild animal. Its behaviour is to be in groups. Many activities are done in groups. There is specific pattern in the activities performed by the deer. It was felt that these activities can be modeled and used to obtain solution of non linear problems. The social activities of deer to form as group motivated us to propose algorithm, called Deer Algorithm. (DA). This proposed Deer algorithm shares many features of the Genetic Algorithms (GA). In this paper we are performing clustering is unsupervised classification on newly proposed algorithms Deer algorithm for the reproduction activity of Deer. Deer algorithms is based on the optimization of some objective function (fitness function). Experimental results are compared with the evolutionary-based clustering algorithm and swarm based clustering algorithms such as PSO, GA, FA, BA. It is shown that the proposed algorithm can achieve the optimal number of clusters and, in most cases, outperforms the other methods on several benchmark datasets in terms of intra cluster and inter cluster distances, robustness, convergence speed and accuracy.