Cuckoo Search Algorithm Inspired by Artificial Bee Colony and Its Application
Cuckoo search algorithm with advanced levy flight strategy, can greatly improve algorithm’s searching ability and increase the diversity of population. But it also has some problems. We improve them in this paper. First, in order to address the randomness
- PDF / 392,918 Bytes
- 12 Pages / 439.37 x 666.142 pts Page_size
- 77 Downloads / 211 Views
ee colony algorithm
1 Introduction Clustering is a common method of data analysis and data mining. Clustering analysis derives from machine learning, pattern recognition, and data mining, etc. When clustering, the data objects are divided into several clusters and the objects in the same cluster have high similarity. Traditional clustering methods are probably divided into the following categories [1]: the method based on hierarchy [2], the method based on density [3–5], the method based on grid and the method based on model [6]. Heuristic intelligent algorithm is one of the research hotspots in the field of computer algorithm design and analysis, which attracted a group of scholars and researchers with its simplicity, distribution, high robustness, and scalability [7]. In recent years, many researchers have used swarm intelligence optimization algorithm for clustering, e.g., T. Niknam proposed a new hybrid evolutionary algorithm to solve nonlinear partition clustering problem, which could find better cluster partition and has the higher robustness [8]; Simulated annealing algorithm and PSO algorithm were applied to cluster respectively [9, 10]; P.S Shelokar put forward an ant colony approach for clustering that the computational simulations revealed very encouraging results in terms of the quality of solution found, the average © Springer International Publishing Switzerland 2016 Y. Tan et al. (Eds.): ICSI 2016, Part I, LNCS 9712, pp. 74–85, 2016. DOI: 10.1007/978-3-319-41000-5_8
Cuckoo Search Algorithm Inspired by Artificial Bee Colony
75
number of function evaluations and the processing time required [11]; Khaled S. Al-Sultan came up with a new algorithm based on Tabu Search(TS) algorithm for clustering, the algorithm performance was better than that of K-means and simulated annealing algorithm [12]; Jie Zhao et al. proposed a new algorithm, which improved cuckoo search algorithm inspired by particle swarm optimization (PSO) algorithm [13]. Cuckoo search (CS) algorithm is a new metaheuristic optimization algorithm developed by Yang and Deb in 2009 [14]. CS algorithm is based on the obligate brood parasitic behavior of some cuckoo species in combination with the Levy flight behavior of birds to solve the optimization problem effectively [15, 16]. Studies have shown that the CS optimization algorithm, with the outstanding search capabilities, effective random search path and few parameters, has been applied to many research areas [17, 18]. But CS algorithm has disadvantage of high randomness and fluctuation in the later stage. To address this problem, we propose a new CS algorithm which combined artificial colony algorithm named as ABC-M-CS and carry it into clustering. The proposed algorithm uses the UCI data sets for simulation, the experimental results reveal very encouraging results in terms of convergence performance and clustering effect.
2 Description of Basic Cuckoo Search Algorithm In nature, the cuckoo looks for suitable birds’ nest randomly, for simplicity in describing the CS algorithm, we now use t
Data Loading...
