Upload
fckw-1
View
222
Download
0
Embed Size (px)
Citation preview
8/16/2019 Nanda and Panda 2013 - A Survey on Nature Inspired Metaheuristic Algorithms for Partitional Clustering
1/18
Review
A survey on nature inspired metaheuristic algorithmsfor partitional clustering
Satyasai Jagannath Nanda a,n, Ganapati Panda b
a Department of Electronics and Communication Engineering, Malaviya National Institute of Technology Jaipur, Rajasthan 302017, Indiab School of Electrical Sciences, Indian Institute of Technology Bhubaneswar, Odisha 751013, India
a r t i c l e i n f o
Article history:
Received 10 October 2012Received in revised form
23 August 2013
Accepted 20 November 2013
Keywords:
Partitional clustering
Nature inspired metaheuristics
Evolutionary algorithms
Swarm intelligence
Multi-objective Clustering
a b s t r a c t
The partitional clustering concept started with K-means algorithm which was published in 1957. Since
then many classical partitional clustering algorithms have been reported based on gradient descentapproach. The 1990 kick started a new era in cluster analysis with the application of nature inspired
metaheuristics. After initial formulation nearly two decades have passed and researchers have developed
numerous new algorithms in this eld. This paper embodies an up-to-date review of all major nature
inspired metaheuristic algorithms employed till date for partitional clustering. Further, key issues
involved during formulation of various metaheuristics as a clustering problem and major application
areas are discussed.
& 2014 Published by Elsevier B.V.
Contents
1. Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2. Single objective nature inspired metaheuristics in partitional clustering. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.1. Problem formulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.2. Historical developments in nature inspired metaheuristics for partitional clustering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.2.1. Evolutionary algorithms in partitional clustering. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.2.2. Physical algorithms in partitional clustering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.2.3. Swarm Intelligence algorithms in partitional clustering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2.4. Bio-inspired algorithms in partitional clustering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2.5. Other nature inspired metaheuristics for partitional clustering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.3. Fitness functions for partitional clustering. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.4. Cluster validity indices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3. Multi-objective algorithms for exible clustering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.1. Problem formulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.2. Historical development in multi-objective algorithms for partitional clustering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.3. Evaluation methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
4. Real life application areas of nature inspired metaheuristics based partitional clustering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
5. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
6. Future research issues. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/swevo
Swarm and Evolutionary Computation
2210-6502/$- see front matter & 2014 Published by Elsevier B.V.
http://dx.doi.org/10.1016/j.swevo.2013.11.003
n Corresponding author.
E-mail addresses: [email protected] (S.J. Nanda), [email protected] (G. Panda).
Please cite this article as: S.J. Nanda, G. Panda, A survey on nature inspired metaheuristic algorithms for partitional clustering, Swarmand Evolutionary Computation (2014), http://dx.doi.org/10.1016/j.swevo.2013.11.003i
Swarm and Evolutionary Computation ∎ (∎∎∎∎) ∎∎∎–∎∎∎
http://www.sciencedirect.com/science/journal/22106502http://www.elsevier.com/locate/swevohttp://dx.doi.org/10.1016/j.swevo.2013.11.003mailto:[email protected]:[email protected]://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003mailto:[email protected]:[email protected]://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://www.elsevier.com/locate/swevohttp://www.sciencedirect.com/science/journal/22106502
8/16/2019 Nanda and Panda 2013 - A Survey on Nature Inspired Metaheuristic Algorithms for Partitional Clustering
2/18
8/16/2019 Nanda and Panda 2013 - A Survey on Nature Inspired Metaheuristic Algorithms for Partitional Clustering
3/18
inspired metaheuristics used in partitional clustering, (2) up-to-
date survey on exible partitional clustering based on multi-
objective metaheuristic algorithms, (3) consolidation of recently
developed cluster validation majors, and (4) exploration of the
new application areas of partitional clustering algorithms.
The paper is organized as follows. Section 2 deals with the
advances in single objective nature inspired metaheuristics for
partitional clustering, which includes recent developments in
algorithm design, tness functions selection and cluster validity
indices used for verication. The multi-objective metaheuristicsused for exible clustering are discussed in Section 3. The real life
application areas of nature inspired partitional clustering are
highlighted in Section 4. Finally the concluding remarks of
investigation made in the survey are presented in Section 5.
A number of issues on innovative future research are presented
in Section 7.
2. Single objective nature inspired metaheuristics in
partitional clustering
2.1. Problem formulation
Given an unleveled dataset Z N D ¼ f z 1D; z 2D;…
; z N Dg repre-senting N patterns, each having D features, partitional approach
aims to cluster the dataset into K groups ðK rN Þ such that
C kaϕ 8 k ¼ 1; 2;…; K ;
C k \ C l ¼ ϕ 8k; l ¼ 1; 2;‥K and ka l; ⋃K
k ¼ 1
C k ¼ Z : ð1Þ
The clustering operation is dependent on the similarity between
elements present in the dataset. If f denotes the tness function
then the clustering task is viewed as an optimization problem as
C kOptimize
f ð Z N D; C kÞ
8 k ¼ 1; 2;…; K ð2Þ
Hence the optimization based clustering task is carried out by
single objective nature inspired metaheuristic algorithms.
2.2. Historical developments in nature inspired metaheuristics
for partitional clustering
In the last two decades a number of nature inspired metaheur-
istics have been proposed in the literature and applied to many
real life applications. In recent years to solve various unsupervised
optimization problems the metaheuristic algorithms are success-
fully used. Present stage for any unsupervised optimization pro-
blem in hand an user can easily pick up a suitable metaheuristic
algorithm for solving the purpose. The solution achieved ensuresoptimality as these population based algorithms explore the entire
search space with the progress in generations.
The basic steps associated with the core algorithms for parti-
tional clustering are listed in Table 2. The recent works on
partitional clustering are outlined in sequence.
2.2.1. Evolutionary algorithms in partitional clustering
The evolutionary algorithms are inspired by Darwin theory of
natural selection which is based on survival of ttest candidate for
a given environment. These algorithms begin with a population
(set of solutions) which tries to survive in an environment
(dened with tness evaluation). The parent population sharestheir properties of adaptation to the environment to the children
with various mechanisms of evolution such as genetic crossover
and mutation. The process continues over a number of generations
(iterative process) till the solutions are found to be most suitable
for the environment. With this concept in mind initially Holland
proposed the Genetic Algorithm (GA) in 1975 [46,47]. It is followed
by development of Evolution Strategies (ES) by Schwefel in 1981
[49–51] and Genetic Programming (GP) by Koza [52] in 1992. Storn
and Price developed another evolutionary concept in 1997 termed
as Differential Evolution (DE) [53]. The books [54,150] on DE,
research work on adaptive DE [55,56] and opposition-based DE
[57,58] made the DE quite popular amongst researchers. The
application of these evolutionary algorithms to partitional cluster-
ing is outlined below.
Table 1
Broad classication of nature inspired metaheuristic algorithms.
Types Single objective Multi-objective
Evolutionary algorithms Genetic Algorithm (GA) [46,47] NSGA II [305,306],
Differential Evolution (DE) [53–58] Multi-objective DE [343]
Genetic Programming (GP) [52] Multi-objective GP [317]
Evolutionary Strategy (ES) [51–139] Multi-objective ES [318]
Granular agent evolutionary algo. [358] SPEA [326], PESA II [325]
Physical algorithms Simulated Annealing (SA) [48] Multi-objective SA [313]
Memetic Algorithm (MA) [167–170] Multi-objective MA [314]
Harmony Search (HS) [173,174] Multi-objective HS [315]
Shuf ed Frog-Leaping algo. (SFL) [179] Multi-objective SFL [316]
Swarm intelligence Ant Colony Opt. (ACO) [62–67] Multi-objective ACO [333]
Particle Swarm Opt. (PSO) [68–72] Multi-objective PSO [307]
Articial Bee Colony (ABC) [73–77] Multi-objective ABC [310]
Fish Swarm algo. (FSA) [254,255] Multi-objective FSA [321]
Bio-inspired algorithms Articial Immune System (AIS) [78–83] Multi-objective AIS [308]
Bacterial Foraging Opt. (BFO) [84,85] Multi-objective BFO [309]
Dendritic Cell algo. [87,88]
Krill herd algo. [356]
Other nature inspired algorithms Cat Swarm Opt. (CSO)[269,270] Multi-objective CSO [311]
Cuckoo Search algo. [272–274] Multi-objective Cuckoo [319]
Firey algo. [275–277] Multi-objective Firey [312]
Invasive Weed Opt. algo. (IWO)[280] Multi-objective IWO [283]Gravitational Search algo. [285,286] Multi-objective GSA [320]
River formation dynamics [357]
Bat algorithm [359,360] Multi-objective Bat [361]
S.J. Nanda, G. Panda / Swarm and Evolutionary Computation ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 3
Please cite this article as: S.J. Nanda, G. Panda, A survey on nature inspired metaheuristic algorithms for partitional clustering, Swarmand Evolutionary Computation (2014), http://dx.doi.org/10.1016/j.swevo.2013.11.003i
http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003
8/16/2019 Nanda and Panda 2013 - A Survey on Nature Inspired Metaheuristic Algorithms for Partitional Clustering
4/18
GA-based approaches: Bezdek et al. [100] initially proposed the useof basic genetic algorithm for partitional clustering. The standard
binary encoding scheme with xed number of cluster centers(k) is used for initialization of chromosomes [100–102]. The repro-
duction operation is carried out using uniform crossover and
cluster-oriented mutation (altering the bits of binary string).
Subsequently integer based encoding of chromosomes is used
by Murthy and Chowdhury [103]. They suggested the use of single
point crossover and Xiaofeng–Palmieri based mutation scheme
[104] for reproduction. However theoretically this mutation may
produce invalid offsprings. Maulik and Bandyopadhyay have
proposed the use of real coded genetic algorithm for partitional
clustering [105]. With real coding the computational complexity is
reduced to O(k) compared to O(nk) associated with integer or
binary encoding. A genetic K-mean algorithm is proposed in [106]
which replaces the crossover phenomenon by the basic search
operation with K-means. Based on this concept Lu et al. havedeveloped fast genetic K-means [108] and incremental genetic k-
means [109] algorithms for gene expression data analysis. Simi-
larly Sheng and Liu [107] have proposed a genetic based hybrid
K-medoid algorithm for accurate clustering of large databases.
All these algorithms are based on a xed number of clusters.
These algorithms work satisfactorily when the suitable number of
partitions for a dataset is known a priori. But in many practical
scenarios the value of K (number of clusters) is unknown to user. The
K value directly affects the partition quality, therefore it is necessary
that the clustering algorithm should explore the number of partitions
along with the process of optimization. Cowgill et al. [110] have
developed a hybrid algorithm COWCLUS which rst uses non-
deterministic genetic algorithm based approach to determine the
good partitions, then used hill-climbing approach to improve these
partitions to produce the nal best partition. Tseng and Yang [111]
have proposed the automatic evolution of clusters with geneticalgorithm. In [112,113] Bandopadhay and Maulik have developed
nonparametric genetic algorithm for automatic selection of number
of partitions K. Based upon this concept a self-adaptive genetic
algorithm for cluster analysis is reported in [125]. Recently a
quantum inspired genetic algorithm for k-means clustering is pro-
posed by Xiao et al. [128] which reports superior performance than
that obtained in [112,113]. The automatic evolution of clusters has
been successfully applied to image classication [113], document
clustering [115], intrusion detection [116], microarray [117] and gene-
expression data analyses [118–120].
In genetic based evolutionary approaches normally a population is
initialized where each individual searches for the optimal weight
vector for all the clusters. Gancarski and Blansche [126,127] devel-
oped co-evolutionary approaches (unlike evolutionary here severalpopulations are employed and each population searches for a local
weight vector for a cluster) based upon Darwinian theory, Lamarck-
ian theory and Baldwin effect for feature weighting in K-means
algorithms. Based upon the three theories they proposed six genetic
approaches for feature weighting in K-means (three based on
evolutionary scheme DE-LKM, LE-LKM and BE-LKM and three co-
evolutionary schemes DC-LKM, LC-LKM and BC-LKM). They reported
that the co-evolutionary based approach for cluster analysis provides
superior performance than the traditional evolutionary based ones.
Intuitively hybrid evolutionary algorithms (formulated by com-
bining the good features of two individual parent processes)
provide superior performance than the conventional parent algo-
rithms. A hybrid of GA and PSO based algorithm is developed in
[129] for order clustering to reduce the surface mount technology
Table 2
Basic steps involved in single objective standard GA, DE, ACO, PSO, ABC, AIS and BFO algorithms for solving partitional clustering problem.
GA ( ( ( Next Generation ( ( (
+ *
Initialize ) Crossover ) Mutation ) Fitness ) Selection ) Cl.
Chromos. O/p
DE ( ( ( Next Generation ( ( (
+ *
Initialize ) Mutation ) Crossover ) Fitness ) Selection ) Cl.
Particles O/p
ACO ( ( ( Next Generation ( ( (
+ *
Initialize ) Fitness ) Update Pheromone ) Drop or ) Short ) Cl.
Ants Intensity Peak Memory O/p
PSO ( ( ( Next Generation ( ( (
+ *
Initialize ) Vel. update ) Compute ) Fitness ) Selection ) Cl.
Particles Pos. update GBst & P Bst O/p
ABC ( ( ( Next Generation ( ( (
+ *
Initialize ) Compute ) Greedy Selection ) Onlooker ) Selection ) Cl.Bees Emp. Bees & Fitness Bees O/p
AIS ( ( ( Next Generation ( ( (
+ *
Initialize ) Fitness ) Clone ) Mutation ) Selection ) Cl.
Immune Cells O/p
BFO ( ( ( Next Generation ( ( (
+ *
Initialize ) Chemotaxis ) Swarming ) Reproduction ) Eliminat. ) Cl.
Bacteria & Dispers. O/p
S.J. Nanda, G. Panda / Swarm and Evolutionary Computation ∎ (∎∎∎∎) ∎∎∎–∎∎∎4
Please cite this article as: S.J. Nanda, G. Panda, A survey on nature inspired metaheuristic algorithms for partitional clustering, Swarmand Evolutionary Computation (2014), http://dx.doi.org/10.1016/j.swevo.2013.11.003i
http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003
8/16/2019 Nanda and Panda 2013 - A Survey on Nature Inspired Metaheuristic Algorithms for Partitional Clustering
5/18
(SMT) setup time. Feng-jie and Ye [130] applied the GA and PSO
based hybrid clustering algorithm for image segmentation of
transmission lines picture to determine the faults. This system is
helpful for remote video monitoring. Hong and Kwong [131]
combined steady-state genetic algorithm and ensemble learning
for cluster analysis. Chaves and Lorenab [132] developed a hybrid
algorithm ‘Clustering Search’ (consisting of GA along with local
search heuristic) to solve capacitated centered clustering problem.
Recently a two stage genetic algorithm was proposed by He et al.[134] for cluster analysis in which two-stage selection and muta-
tion operations are incorporated to enhance the search capability
of the algorithm. The two stage genetic algorithm provides
accurate results compared to agglomerative k-means [133] and
standard genetic k-means algorithms. A grouping genetic algo-
rithm (GGA) is a compact one proposed by Falkenauer [135] to
handle grouping-based problems. The GGA is successfully used for
cluster analysis of benchmark UCI datasets in [136]. Recently Tan
et al. [137] applied the GGA based clustering technique to improve
the spectral ef ciency of OFDMA (orthogonal frequency-division
multiple access) based multicast systems.
ES-based approaches: Babu and Murty [138] developed thepartitional and fuzzy clustering algorithms with ES in 1994. They
have used the minimization of WGSS (within group sum of
squared error) objective function for partitional clustering and
minimization of FCM (fuzzy C-means) objective functions for
fuzzy clustering. The paper by Beyer and Schwefel [139] discusses
the fundamental and recent advancements in partitional cluster-
ing with ES. Hybrid partitional clustering algorithm based on K-
means and ES is developed in [140]. It is observed that the hybrid
algorithms provide better performance than the regular ES on
cluster analysis of benchmark UCI datasets. The ES based parti-
tional clustering has been suitably used for cluster analysis of DNA
microarray database [141]. GP-based approaches: The GP is related to GA, where it auto-
matically generates computer programs, based on the Darwin
principle. Each individual computer program is a solution to
the optimization problem and is encoded in the form of a treecomprising functions and terminals. The GP has been widely
used for supervised classication problem and it is reported
that the trees generated by GP have capability to separate
regions with varieties of shapes [142–144]. Falco et al. [145,146]
developed the partitional clustering algorithm based on GP. The
algorithm starts with a population of program trees generated
at random. The algorithm determines the optimal number of
clusters by selecting a variable number of trees per individual.
The user has to provide a parameter that directly inuences the
number of clusters present in the dataset. The trees undergotness evaluation and those having higher tness have the
higher probability to serve as parents for next generation. The
genetic operators like crossover and mutation are applied on
the parent trees to generate offspring. The process continuestill a predened stopping criteria corresponding to the optimal
cluster partition get satised. Boric et al. [147] modied the GP
based partitional clustering with an information theoretic
tness measure which can determine arbitrary shape clusters
present in the dataset. DE-based approaches: The book on Metaheuristic Clustering by
Das et al. [2] in 2009 discusses the fundamental as well as the
advances in DE approaches for cluster analysis [150]. In case of
DE based clustering the individual target solutions (which
combines to create a population P ) are taken as parameter
vectors or genomes. Each target vector xi ¼ ½mi1; mi2;‥;
mik;‥; miK , where mik is the centroid of cluster c ik and K
represents the number of clusters. Then DE employs the
mutation operation to produce a mutant vector vi. The ve
most commonly used mutation strategies are
v1;i ¼ xr 1 ;i þ F ð xr 2 ;i xr 3 ;iÞ
v2;i ¼ xbest þ F ð xr 1 ;i xr 2;iÞ
v3;i ¼ xi þF ð xbest ;i xiÞ þF ð xr 1 ;i xr 2;iÞ
v4;i ¼ xbest þ F ð xr 1 ;i xr 2;iÞ þF ð xr 3;i xr 4 ;iÞ
v5;i ¼ xr 1 ;i þ F ð xr 2 ;i xr 3 ;iÞ þF ð xr 4 ;i xr 5;iÞ ð3Þ
where i varies from 1 to P and r 1; r 2; r 3; r 4; r 5 are mutually
exclusive integers randomly generated within the range [1, P ].The scale factor F is a control parameter used for amplication
of difference vector, normally lies in range [0,2]. Then a cross-
over operation is applied to each pair of the target vector xi and
its corresponding mutant vector vi to obtain a trial vector ui as
ui ¼vi if ðrand1rCRÞ or ði ¼ irandÞ
xi Otherwise; 8 i ¼ 1; 2;‥; P
( ð4Þ
where rand1 is a random number in [0,1]. The crossover rate CR
is an user dened constant in the range [0,1]. The irand is a
randomly chosen integer in the range [1, P ]. The tness of all
target vector xi and trail vector ui is evaluated using one of the
tness functions dened in Table 3. Then the population for
next generation is given by
xt þ 1i ¼ut i if f ðu
t i Þr f ð x
t i Þ
xt i Otherwise; 8 i ¼ 1; 2;‥; P
( ð5Þ
where t is the number of generation. The algorithm run for
certain number of generations till the algorithm converges and
the optimum clusters are achieved.
The benchmark research article on DE based automatic clustering
[16] is published in 2008 by Das et al. Prior to that the DE based
framework introduced for partitional clustering by Paterlini and
Krink [148,149] is worth mentioning. Further research work by Das
et al. deals with hybridization of Kernel-based clustering with DE
[151] and application of DE based clustering algorithms to image
pixel clustering [152]. Subsequently various hybrid algorithms based
on DE are developed by several researchers which include DE-K
means by Kwedlo [153], DE-K harmonic means by Tian et al. [154],
DE-possibilistic clustering by Hu et al. [71]. These algorithms have
been successfully applied to image classication [156], document
clustering [157] and node selection in mobile networks [158].
2.2.2. Physical algorithms in partitional clustering
The physical algorithms are inspired by physical processes such as
heating and cooling of materials (Simulated Annealing given by
Kirkpatrick et al. in 1983 [48]), discrete cultural information which is
treated as in between genetic and culture evolution (Memetic
algorithm by Moscato [167] in 1989), harmony of music played by
musicians (Harmony Search by Geem et al. [173] in 2001) and
cultural behavior of frogs (Shuf ed frog-leaping algorithm by Eusuff et al. [179] in 2006). These algorithms have been applied to solve
partitional clustering problem as briey explained in sequence:
Simulated Annealing (SA) based approaches: Selim and Alsultanrst developed the SA based partitional clustering in 1991
[159]. Then in 1992 Brown and Huntley [160] applied the SA
based partitional clustering algorithm to solve multi-sensor
fusion problem. The clustering algorithm begins with an initial
solution ‘ x’ (cluster centroids) having a large initial temperature‘T ’. The tness of the initial solution ‘ f ( x)’ (computed with any
function from Table 3) represents the internal energy of the
system. The heuristic algorithm moves to a new solution ‘ x’
(selected from its neighborhoods of a state) or remain in the
old state ‘ x’ depending upon a acceptance probability function
S.J. Nanda, G. Panda / Swarm and Evolutionary Computation ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 5
Please cite this article as: S.J. Nanda, G. Panda, A survey on nature inspired metaheuristic algorithms for partitional clustering, Swarmand Evolutionary Computation (2014), http://dx.doi.org/10.1016/j.swevo.2013.11.003i
http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003
8/16/2019 Nanda and Panda 2013 - A Survey on Nature Inspired Metaheuristic Algorithms for Partitional Clustering
6/18
given by
P ðaccept Þ ¼ exp f ð xÞ f ð x0Þ
T
ð6Þ
where f ( x) is the energy and T is the temperature of the present
state. The probability function P (accept ) is positive when f ð x0Þis lower than f ( x) which represents that the smaller energy
solutions are better than those with a greater energy. The
temperature ‘T ’ plays a crucial role in controlling the evolution
of the state with the cooling process of the system. The
algorithm continues either for a xed number of iterations or
until a state with minimum energy is found (global solution
corresponds to optimal cluster partition).
The basic SA has been suitably combined with K-means [161]
and K-harmonic means [162] to develop hybrid algorithms which
provide superior performance in accurately clustering the UCI
datasets. A GA and SA based hybrid clustering algorithm is
developed in [164] to solve the dynamic topology management
and energy conservation problem in mobile ad hoc network.
Lu et al. [165] developed a fast simulated annealing based cluster-
ing approach by combining multiple clusterings based on different
agreement measures between partitions. Recently the SA based
clustering is applied to group the suppliers for effective manage-
ment and to fulll the demands of customers (i.e. to build a good
supply chain management system) [166].
Memetic Algorithm-based approaches: The recent survey articlesby Chen et al. [169,170] highlight the recent advances in theory
and application areas of memetic algorithm. This algorithm is
used by Merz [168] to perform cluster analysis on gene
expression proles using minimization of the sum-of-squares
as tness measure. It begins with a population which under-
goes a global search (exploration of various areas of the search
space), combined with an individual solution improvement
(performed by a local search heuristic to provide local rene-
ments). A balance mechanism is carried out with the local and
global mechanisms to ensure that the system does not achieve
premature convergence to a local solution as well as it does
not consume more computational resources for achieving the
Table 3
Similarity functions f ðÞ used by the single objective nature inspired metaheuristic algorithms for cluster analysis. Considering dataset Z N D ¼ f z 1D; z 2D;…; z N Dg to be
divided into K clusters with valid partitions C k as per (1).
S imilarity fun. Characterist ics Details
Medoid distance Explanation Minimization of sum of Distance between objects and medoids of dataset
RepresentationF 1 ¼ ∑
N i ¼ 1 jAf1;‥K g
min
dð z i ; m j Þ where medoids fm1; m2;…; mkg Z , d is any distance
Used in Lucasius et al. [121], Castro and Murray [122], Sheng and Liu [107]
Centroid distance Explanation Minimization of sum of squared Euclidean distance of objects from respective cluster meansRepresentation F 2 ¼ ∑
K j ¼ 1∑ z i Ac j J z i μ j J
2, with μ j is the mean of c j
Used in Maulik and Bandyopadhyay [105], Zhang and Cao [207], Murthy and Chowdhury [103]
Distortion distance Explanation Minimization of intraclus- ter diversity
Representation F 3 ¼ F 2=ðN DÞ
Used i n K ri shna a nd Mu rty [106], Lu et al. [108,109], Franti et al. [124], Kivijarvi et al. [125]
Variance ratio criterion Explanation It is the ratio of between cluster (B) and pooled within cluster (W ) covariance matrices. The VRC should be maximized
RepresentationF 4 ¼ VRC ¼
trace B=ðK 1Þ
trace W =ðN K Þ
Used in Cowgill et al. [110], Casillas et al. [115]
Intra- and inter-clust. distance Explanation Difference between inter-cluster to intra-cluster dist.
Representation F 5 ¼ ∑K i ¼ 1 Dinter ðc iÞw Dintraðc i Þ, w is a parameter
Used in Tseng and Yang [111]
Dunn0s index Explanation Dunn0 s index to be maximized for optimal partition
RepresentationF 6 ¼ DI ðK Þ ¼ iAK
min
jAK ; jaimin
δ ðc i ;c j ÞkAK
max
f Δðc k Þg
, where δ ðc i ; c jÞ ¼ minfdð z i; z j Þ : z iAc i ; z jAc j g, Δðc k Þ ¼ maxfdð z i; z j Þ : z i ; z jAc ig,
d is the distance
Used in Dunn [293], Zhang and Cao [207]
Davis–Bouldin (DB) index Explanation Ratio of sum of within cluster scatter to between cluster separation. DB index to be minimized
RepresentationF 7 ¼ DBðK Þ ¼
1k∑K i ¼ 1 Ri;qt , where Ri;qt ¼ jAK ; ja i
max S i;q þS j;qdij;t
The ith cluster scatter S i;q ¼ 1
N i∑ z A c i J z μi J
qh i1=q
, where N i and μ i are the number of elements and center
of c i respectively. The separation distance between ith and jth cluster is dij;t ¼ ½∑Dd ¼ 1 j μi;d μ j;dj
t 1=t
Used i n D avis a nd B ou ldi n [291], Cole [123], Das et al. [16], Bandyopadhyay and Maulik [113], Agustin-Blas et al. [136]
C S measure Expl ana tion C S M ea su re i s to b e min imiz ed for optimal p ar ti tioni ng
Representation
F 8 ¼ CS ðK Þ ¼∑K i ¼ 1
1N i∑ z j Ac i z qAc i
maxdð z j ; z qÞ h i
∑K i ¼ 1 jAK ; jai
min
fdðmi ; m j Þg
, with centroid mi ¼ 1N i∑ z j Ac i z j , where N i is number of elements in c iUsed in Chou et al. [292], Das et al. [16]
Silhouette Explanation Higher silhouette provides better assignment of elements
RepresentationF 9 ¼ ∑
N i ¼ 1
S ð z i Þ
N where element z iA A, with A; B c k
S ð z i Þ ¼ bð z i Þ að z iÞ
maxfað z i Þ; bð z i Þg, Silhouette range: 1rS ð z i Þr1,
að z i Þ is the average dissimilarity of z i to other elements of A,
neighbor dissimilarity bð z iÞ ¼ min dissð z i ; BÞ, AaB
Used in Kaufman and Rousseeuw [294], Hruschka et al. [3,118]
S.J. Nanda, G. Panda / Swarm and Evolutionary Computation ∎ (∎∎∎∎) ∎∎∎–∎∎∎6
Please cite this article as: S.J. Nanda, G. Panda, A survey on nature inspired metaheuristic algorithms for partitional clustering, Swarmand Evolutionary Computation (2014), http://dx.doi.org/10.1016/j.swevo.2013.11.003i
http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003
8/16/2019 Nanda and Panda 2013 - A Survey on Nature Inspired Metaheuristic Algorithms for Partitional Clustering
7/18
solution. The memetic based partitional clustering algorithm
has been applied for energy ef cient clustering of nodes in
wireless sensor networks [171] and segmentation of natural
and remote sensing images [172]. Harmony Search: The Harmony search algorithm becomes
popular after Lee and Geem [174] applied it for various
engineering optimization problems. Mahdavi et al. developed
the Harmony search based partitional algorithm for web page
clustering [175,176]. The algorithm is inspired by the harmonyplayed by the musicians. Here each musician represents a
decision variable which denotes solution of the problem. The
musicians try to match harmony with respect to time by
incorporating variation and improvisations in the pitch played
by him. The variation in pitch is given by x 0 ¼ x þ PB ε, wherePB is the pitch bandwidth which is an user dened parameter
to control the amount of change and ε is a random number inthe range [ 1,1]. This variation is reected in the form
of improvement in the cost function to achieve the global
solution. Mahdavi and Abolhassani have also formulated
a hybrid Harmony K-means algorithm for document clustering
[177]. The clustering algorithm [178] has been suitably
applied for designing clustering protocols for wireless sensor
networks. Shuf ed frog-leaping algorithm (SFL): The SFL algorithm mimics
the nature of frogs in the memeplexes. The algorithm is used to
solve partitional clustering problem [180] and has been
reported to yield better solutions than ACO, simulated anneal-
ing, genetic k-means [106] approaches on several synthetic and
real life datasets. The initial population consists of a set of frogs
(solutions) which is grouped into subsets known as meme-
plexes. The frogs which belong to different memeplexes are
assumed to be of different cultures and are allowed to perform
local search. So within each memeplexes each individual frog
shares its ideas with other frogs and thus the group evolves
with new ideas (with memetic evolution). After a pre-dened
number of steps (with memetic evolution), the ideas are shared
among the memeplexes using a shuf ing process. The local
(memetic) and global searches (shuf ing process) continue tillthe optimal tness (accurate clusters) is achieved. The cluster-
ing algorithm based on SFL has been used for color image
segmentation [181] and web0s text mining [182].
2.2.3. Swarm Intelligence algorithms in partitional clustering
Swarm intelligence is the group of natural metaheuristics
inspired by the ‘collective intelligence’. The collective intelligence
is built up through a population of homogeneous agents interact-
ing with each other and with their environment. Example of such
intelligence is found among colonies of ants, ocks of birds,
schools of sh, etc. The books [59–61] highlight the fundamentals
and developments in swarm intelligence algorithms for solvingnumerous real life optimization problems. The major such algo-
rithms include: Ant colony optimization (ACO) by Dorigo [62] in
1992, Particle swarm optimization (PSO) by Kennedy and Eberhart
in 1995 [68,69], Articial bee colony (ABC) algorithm by Karaboga
and Basturk in 2006 [73], Fish Swarm Algorithm (FSA) by Li et al.
in 2002 [254,255]. Application of these algorithms to solve parti-
tional clustering problems is outlined in sequence
ACO-based approaches: The ACO algorithm is inspired by antsbehavior in determining the optimal path from the nest to the
food source. The algorithm becomes popular after Dorigo et al.
work was standardized in IEEE [63–65]. With the progress
of time Dorigo0s book on ACO [66] and survey paper [67]
are heavily cited by the researchers and scientists in this eld.
The cluster analysis algorithms based on ACO follow either of
the two fundamental natures of real life ants.
First one is based on ants foraging behavior for determining the
food source. Initially ants wander randomly for food in the surround-
ing regions of nest. An ant0s movement is observed by the neighboring
ants with the pheromone intensity it lays down while searching for
food. Once a food source is found the pheromone intensity of the path
increases due to the movement of ant from source to nest and otherants instead of searching at random, they follow the trail. With the
progress in time the pheromone intensity starts to evaporate and
reduce its attraction. The amount of time taken for an ant to travel to
food source and back to the nest is directly proportional to the
quantity of pheromone evaporation. So with time an optimal shortest
path is achieved to maintain the high pheromone intensity. With this
concept the cluster analysis is formulated as an optimization problem
and solved using ACO to obtain the optimal partitions in [183,184]. A
constrained ACO (C-ACO) [185] was proposed to handle arbitrary
shaped clusters and outliers present in the data. Then adaptive ACO
was proposed by several researchers [186–188] to improve the
convergence rate and to determine the optimal number of clusters.
A variant of ACO, known as APC (aggregation pheromone density-
based clustering) algorithm is proposed by Ghosh et al. [189,190]. The
beauty of APC is updation of the pheromone matrix which is helpful to
avoid the convergence of solutions to a local optima.
The second one imitates the ants behavior of grouping dead
bodies. Ants work together to deposit more dead bodies in their
nest and group them with respect to their size. This grouping
property of ants is rst coded in the form of algorithm for data
clustering (LF algorithm) by Lumer and Faieta [191]. The basic LF
algorithm was followed and improved by several researchers
[192,193]. Yang et al. [194,195] proposed the use of multi-ant
colonies algorithm for clustering. In this concept, the algorithm
consists of several independent ant colonies (each having a queen
ant). The moving speed of ants and parameters of the probability
conversion function in different colonies differ from each other.
Each colony produces a clustering result in parallel and sends it to
the queen ant agent. A hypergraph model (through queen ants) isused to combine all the parallel colonies.
Handel et al. published a number of articles on ACO [196–199]
which are extensively cited by the researchers. They have incor-
porated robustness in the standard LF algorithm (known as ACA)
and applied it for document retrieval [196]. The performance of
these methods have been compared with that obtained by ant-
based clustering with K-means, average link and 1DSOM in [197].
They have suggested an improved ACO in [198] which incorporates
adaptive and heterogeneous ants for better exploration of search
space. In the survey article [199] both the approaches of ACO along
with other swarm based clustering approaches (like bird ocking
algorithm and PSO) have been dealt. A modied version of ACA
(known as ACAM) is proposed by Boryczka et al. [200] which has
been shown to outperform ACA [196] in terms of accuracy (testedwith ve cluster validation measures). Recently an automatic
clustering based on ant dynamics is proposed in [201], which
can detect arbitrary shape clusters (both convex and/or non-
convex). Another algorithm known as chaotic ant swarm (CAS)
proposed by Wan et al. [202] provides optimal partitions irrespec-
tive of cluster size and density.
A number of hybrid algorithms based on ants are available in
the literature. Initially Kuo et al. [203] have proposed ants based K-
means algorithm, which is subsequently improved by hybridiza-
tion of ACO, self-organizing maps(SOM) and k-means in [204].
Further, Jiang et al. have developed new hybrid clustering algo-
rithms by combining the ACO with K-harmonic means algorithm
in [205] and DSBCAN algorithm in [206]. Recently Zhang and Cao
[207] have suggested a new one by integrating ACO with kernel
S.J. Nanda, G. Panda / Swarm and Evolutionary Computation ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 7
Please cite this article as: S.J. Nanda, G. Panda, A survey on nature inspired metaheuristic algorithms for partitional clustering, Swarmand Evolutionary Computation (2014), http://dx.doi.org/10.1016/j.swevo.2013.11.003i
http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003
8/16/2019 Nanda and Panda 2013 - A Survey on Nature Inspired Metaheuristic Algorithms for Partitional Clustering
8/18
principal component analysis (KPCA). Here the KPCA is applied on
the dataset to compute ef cient features and then ant based
clustering is performed in the feature space (instead of the input
space). A multiple cluster detection algorithm based on spatial
scan statistic and ACO is reported in [208]. It is observed that these
hybrid algorithms exhibit performances superior to that of the
individual algorithms in terms of ef ciency and clustering quality.
The ant based clustering algorithms nd applications to web
mining [209], test mining [188], texture segmentation [210],intrusion detection [211,212], high dimensional data analysis
[213], long-term electrocardiogram processing [214] and gene
expression data analysis [215].
PSO-based approaches: The PSO is based on the swarmingbehavior of particles searching for food in a collaborative
manner. The algorithm has become popular among the
researchers [70–72] due to its simple form for implementation,
easier selection of parameters and faster convergence rate.
The cluster analysis using PSO was proposed by Omran et al.
[219] for image clustering. Then van der Merwe and Engelhrecht
[220] applied it for cluster analysis of arbitrary datasets. The
algorithm in its basic form for cluster analysis consists of a swarmin a D dimensional search space in which each particle 0s position
xi ¼ ½mi1; mi2;‥; mik;‥; miK consists of K cluster centroid vectors.
The mik is the centroid of cluster c ik. The position of ith particle is
associated with a velocity V i ¼ ½vi1; vi2;‥; viK , where vi1; vi2 are
initialized as random numbers in the search range. Then the
tness of particles is evaluated with a suitable tness function f ð:Þ dened in Table 2. Based on the tness values the best previous
positions achieved by the particles represent the local solutions
given by P i ¼ ½ pi1; pi2;‥; piK . For the initial run P i ¼ xi. The global
solution is the best position achieved by the swarm in a generation
given by P g ¼ ½ p g 1; p g 2;‥; p gt , where t is the number of generation.
The cluster centroid positions are updated with the velocity and
position update of the particles given by
vikðt þ 1Þ ¼ w vikðt Þ þc 1 r 1 ð pikðt Þ xikðt ÞÞþc 2 r 2 ð p g ðt Þ xikðt ÞÞ ð7Þ
xikðt þ1Þ ¼ xikðt Þ þ vikðt þ1Þ ð8Þ
where r 1 and r 2 represent random numbers between [0, 1], w is
the inertia weight which is taken as 0.4. The c 1 and c 2 are
acceleration constants taken as 2.05. The updation process con-
tinues till the number of data points which belongs to each cluster
remains constant for certain generations.
A number of variants of PSO based clustering algorithms have
been reported by researchers in the last couple of years. Cohen and
Castro [221] have proposed a particle swarm clustering (PSC)
algorithm in which the particle0s velocity update is inuenced by
particle0s previous position along with a cognitive term, social
term and self-organizing term. These terms are helpful to guidethe particle for better solutions and to avoid local stagnation.
A combinatorial particle swarm optimization (CPSO) based parti-
tional clustering is proposed in [222] for solving multi-mode
resource-constrained project scheduling problem. Chuang et al.
[223] developed a chaotic PSO which replaces the convergence
parameters like w, c 1, c 2, r 1, r 2 with chaotic operators. These new
operators incorporates ergodic, irregular, and stochastic properties
of chaos in PSO to improve its convergence. A selective particle
regeneration based PSO (SRPSO) and a combination of it with K-
means (KSRPSO) are proposed in [224] for partitional clustering.
Both algorithms provide faster convergence than PSO and
K-means, due to particle regeneration operation that enables
better exploration of search space. Sun et al. [225] proposed a
quantum-behaved PSO (QPSO) algorithm for cluster analysis of
gene expression database. Recently a new PSO based partitional
clustering algorithm is developed by Cura et al. [226] to handle
unknown number of clusters.
The hybrid algorithm based on K-means and PSO is proposed
by van der Merwe and Engelhrecht [220] in 2003. The PSO has
been suitably combined with K-harmonic means [227] and rough
set theory [228] to produce hybrid algorithms for partitional
clustering. Du et al. have formulated a DK algorithm [229] by
hybridizing particle-pair optimizer (PPO) algorithm (a variation onthe traditional PSO) with K-means for microarray data cluster
analysis. The DK algorithm is reported to be more accurate and
robust than K-means and Fuzzy K-means(FKM) algorithm. Zhang
et al. [230] combined PSO with possibilistic C-means(PCM) for
image segmentation which provides superior performance than
fuzzy C-means(FCM) algorithm. Another ef cient approach based
on PSO, ACO and K-means for cluster analysis is reported in [231].
Recently several researchers have produced new hybrid evolu-
tionary clustering algorithms by suitably combining PSO with
differential evolution [232], genetic algorithm [233], immune
algorithms [234,235] and simulated annealing [236]. These hybrid
algorithms provide superior performance than the individual
traditional evolutionary algorithms in terms of ef ciency, robust-
ness and clustering accuracy.
The PSO based clustering algorithms have been effectively used
in several real life applications including node clustering in
wireless sensor network (WSN) to enhance lifetime of sensors
and coverage area [17], energy balanced cluster routing in WSN
[237], clustering in mobile ad hoc networks to determine the
cluster heads which becomes responsible for aggregating the
topology information [238], cluster analysis of stock market data
for portfolio management [27], grouping for security assessment
in power systems [239], gene expression data analysis [240], color
image segmentation [241], clustering for manufacturing cell
design [242], image clustering [243], document clustering [244],
cluster analysis of web usage data [245] and network anomaly
detection [246].
ABC-based approaches: The ABC algorithm mimics the foragingbehavior of honey bee swarm. The algorithm has become
popular after a sequence of publication made by Karaboga
et al. [74–77]. Recently the ABC algorithm is used for cluster
analysis by several researchers like Zhang et al. [247], Zou et al.
[248], Fathian et al. [249] and Karaboga et al. [250].
The clustering algorithm based on ABC begins with initialization of
bee population with randomly selected cluster centroids in the
dataset. The initial population is categorized into two parts: employed
bees and the onlookers. The employed bees are always associated with
a food source. The food source represents the quality of the solution
(in terms of tness) to the problem and to be optimized. An employed
bee modies its position (i.e. determines a new food source) depend-
ing upon local information and the tness value of new source. If thetness value of new source is more than the previous one than the
employed bee memorizes the new position and forgets the old one.
After all employed bees complete the search, they share the informa-
tion on food sources and their position with the onlooker bees on the
dance area. Then the onlooker bees are assigned as employed bees
based on a probability which is related to the tness of the food
source. These bees now update their position and share their
information. Every bee colony has scout bees which do random search
in the environment surrounding the next to discover new food
sources. This process is helpful for exploration in the search space
and to avoid the solutions being trapped into a local food source
(optima). The clustering algorithm based on ABC has been suitably
applied for solving network routings [251] and sensor deployment
problems [252] in wireless sensor networks. Recently a hybrid
S.J. Nanda, G. Panda / Swarm and Evolutionary Computation ∎ (∎∎∎∎) ∎∎∎–∎∎∎8
Please cite this article as: S.J. Nanda, G. Panda, A survey on nature inspired metaheuristic algorithms for partitional clustering, Swarmand Evolutionary Computation (2014), http://dx.doi.org/10.1016/j.swevo.2013.11.003i
http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003
8/16/2019 Nanda and Panda 2013 - A Survey on Nature Inspired Metaheuristic Algorithms for Partitional Clustering
9/18
clustering algorithm HABC is proposed by Yan et al. [253] by
incorporating crossover operation of GA in ABC which provides
superior performance than that obtained by each of PSO, CPSO, GA,
ABC and K-means algorithm.
Fish Swarm Algorithm (FSA): The FSA algorithm is derived fromthe schooling behavior of sh. Cheng et al. [256] applied the
FSA for cluster analysis. The algorithm operates by mimicking
three important behavior of natural sh: searching behavior(tendency of sh to look at food), swarming behavior (sh
assembles in swarms to minimize danger) and following
behavior (when a sh identify food source, its neighboring
individuals follow based on sh0s visual power). Tsai and Lin
[257] have reported improved solution provided by FSA com-
pared to PSO for several optimization problems.
2.2.4. Bio-inspired algorithms in partitional clustering
The bio-inspired, short form of biologically inspired algorithms
comprise natural metaheuristics derived from living phenomena
and behavior of biological organisms. The intelligence derived
with bio-inspired algorithms are decentralized, distributed, self-
organizing and adaptive in nature (under uncertain environ-
ments). The major algorithms in this eld include Articialimmune systems (AIS) [78–83], Bacterial foraging optimization
(BFO) [84–86], Dendritic cell algorithm [87,88] and Krill herd
algorithm [356]. The usage of these algorithms to ef ciently solve
partitional clustering problem is highlighted for each case:
AIS-based approaches: The books by Dasgupta [78], Charsto andTimmis [79] provide the fundamental concepts on articial
immune system for computing and its potential applications.
The four core models developed by mimicking the principle of
biological immune system include: negative selection algorithm,
clonal selection algorithm, immune network model and danger
theory. Among these four the clonal selection principle by
Charsto and Zuben [80] has becomes popular for machine
learning and optimization purposes. The recent articles byDasgupta et al. [81,82] and thesis by Nanda [83] highlight the
major advances in the theory and applications of AIS.
Initially Nasraoui et al. [258] developed an AIS based model for
dynamic unsupervised learning. Then the clonal selection algo-
rithm [259,260] has been effectively used for cluster analysis.
In this algorithm the immune cells (they combine to form a
population which is responsible to protect the body against
infection) are initialized with K cluster centroid vectors. When
an antigen (foreign element) invades the body; number of anti-
bodies (immune cells) that recognize these antigens survive
(based on the best tness value). These immune cells undergo
clonal reproduction (new immune cells are produced which are
copies of ef cient parent cells). Then a portion of cloned popula-tion undergoes a mutation mechanism (somatic hypermutation).
The mutation mechanism is responsible to diversify the solutions
in the search space, thus avoids the cells to be trapped in the local
optima. The best particles among the mutated and cloned ones are
kept as the parents for next generation. The algorithm runs for a
xed number of generations (user dened) till the convergence is
achieved and a optimal number of clusters are obtained.
Li and Tan [261] rst developed the hybrid clustering algorithm
based on AIS by combining it with support vector machine (SVM).
Then an immune K-means algorithm is developed in [262] which
is based on the negative selection principle. Nanda et al. [234]
developed an Immunized PSO (IPSO) algorithm in which the global
best particle is cloned and mutated after the velocity and position
update to enhance the particles search in an focused manner. In a
recent work the IPSO has been suitably employed for partitional
clustering task [234]. Graaff and Engelbrecht [263] initially devel-
oped a local network neighborhood clustering method based on
AIS. Later on they have formulated the immune based algorithm
for cluster analysis under uncertain environments [264].
BFO-based approaches: Passino [84] proposed the bacterialforaging optimization (BFO) algorithm in 2002 which imitates
the foraging strategies of E. coli bacteria for nding food. An E.coli bacterium can search for food in its surrounding by two
types of movements: run or tumble. These movements are
possible with the help of agella (singular, agellum) that enable
the bacterium to swim. If the agella move counterclockwise,
their effects accumulate in the form of a bundle which pushes
the bacterium to move forward in one direction (run). When the
agella rotate clockwise, each agellum separates themselves
from the others and the bacterium tumbles (it does not have any
set direction for movement and there is almost no displace-
ment). The bacterium alternates between these two modes of
operation throughout its entire lifetime. After the initial devel-
opment by Passino the algorithm gradually has become popular
due to its capability to provide good solution in dynamic [85]
and multi-modal [86] environments.
Literature review indicates that this algorithm has recently
being applied to cluster analysis [265, 266]. The basic clustering
algorithm based upon the BFO consists of four fundamental steps:
chemotaxis, swarming, reproduction, elimination and dispersal.
The initial solution space is created by assigning the bacteria
positions as the randomly chosen cluster centroids in the dataset.
Then the chemotaxis process denes the movement of bacteria,
which represents either a tumble followed by a tumble or tumble
followed by a run. The detailed mathematical expression in
chemotaxis for the movement of bacteria (i.e. cluster head) is
dened in [84]. The swarming operation represents the cell-to-cell
signaling scheme of bacteria via an attractant. The clustering task
can be also performed satisfactorily without the swarmingscheme (which involves high computational complexity and thus
eliminated in [267]). After performing a xed number of chemo-
taxis loops the reproduction is carried out where the population is
sorted with respect to the tness value. The rst half of the bacteria
is retained and the second half (i.e. least healthy bacteria) is
allowed to die. Each of the healthiest bacteria splits into two
bacteria, which are placed at the same location. In order to prevent
the bacteria from being trapped into local optima the elimination
and dispersal phases are carried out. Here a bacterium is chosen
according to a preset probability and is allowed to disperse
(i.e move to another random position). The dispersal at times
becomes useful as it may place bacteria near good food sources (i.e.
optimal cluster partitions). The BFO based clustering algorithm has
been successfully applied to deploying sensor nodes in wirelesssensor network to enhance the coverage and connectivity [268].
2.2.5. Other nature inspired metaheuristics for partitional clustering
Cat Swarm Optimization (CSO) – The CSO algorithm is proposedby Chu and Tsai [269,270] by observing the natural hunting
skill of cats. Santosa et al. [271] used CSO based clustering to
classify benchmark UCI datasets [354]. The algorithm deter-
mines the optimal solution based on two modes of operation
cats: seeking mode (represents global search technique which
mimics the resting position of cats with slow movement) and
tracing mode (local search technique which reects the rapid
chase of cat behind the target). Recently Pradhan et al. [20]
S.J. Nanda, G. Panda / Swarm and Evolutionary Computation ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 9
Please cite this article as: S.J. Nanda, G. Panda, A survey on nature inspired metaheuristic algorithms for partitional clustering, Swarmand Evolutionary Computation (2014), http://dx.doi.org/10.1016/j.swevo.2013.11.003i
http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003
8/16/2019 Nanda and Panda 2013 - A Survey on Nature Inspired Metaheuristic Algorithms for Partitional Clustering
10/18
applied the multi-objective CSO algorithm for optimal deploy-
ment of sensor nodes in wireless sensor networks. Cuckoo Search Algorithm – The cuckoo search algorithm is
developed by Yang and Deb [272] in 2009. The algorithm
mimics the breeding behavior of cuckoos (to lay their eggs
in the nests of other birds). Three basic operations associated
are: (i) every cuckoo lays one egg at a time, and dumps its egg
in randomly selected nest in the environment, (ii) the nests
with good quality of eggs will remain for next generations, (iii)the number of host bird nests is xed, and the egg laid by a
cuckoo is identied by the host bird depending on a probability
in the range [0, 1] (under such situation, the host bird can
either destroy the egg or destroy the present nest and build a
new one). Goel et al. [274] have formulated the cuckoo search
based clustering algorithm and applied it for extraction of
water body information from remote sensing satellite images. Fire y algorithm – The algorithm is proposed by Yang [275–277]
observing the rhythmic ashes of reies. Senthilnath et al.
[278] applied the algorithm for cluster analysis of UCI datasets.
The algorithm follows three rules based upon the glowing
nature of reies: (i) all reies are unisex and each rey is
attracted towards other reies regardless of their sex; (ii) the
attraction is proportional to their brightness. Therefore between
any two ashing reies, the less brighter one will move
towards the brighter one. As the attraction is proportional to
the brightness, both decrease with the increase in distance
between reies. In the surrounding if there is no brighter one
than a particular rey, then it has to move randomly; (iii) the
brightness of a rey is determined by the nature of objective
function. Initially at the beginning of clustering algorithm, all the
reies are randomly dispersed across the entire search space.
Then the algorithm determines the optimal partitions based on
two phases: (i) variation of light intensity: the brightness of a
rey at current position is reected on its tness value, (ii)
movement towards attractive rey: the rey changes its
position by observing the light intensity of adjacent reies.
Hassanzadeh et al. [279] have successfully applied the rey
clustering algorithm for image segmentation. Invasive Weed Optimization Algorithm (IWO) – The IWO is
proposed by Mehrabian and Lucas [280] by following the
colonization of weeds. The weeds reproduce their seeds spread
over a special area and grow to new plants in order to nd the
optimized position. The automatic clustering algorithm based
upon IWO is formulated by Chowdhury et al. [281]. The
algorithm is based upon four basic steps: (i) initialization of
the weeds in the whole search space, (ii) reproduction of the
weeds, (iii) distribution of the seeds, (iv) competitive exclusion
of the weeds (tter weeds produce more seeds). Su et al. [282]
applied the algorithm for image clustering. The multi-objective
IWO is proposed by Kundu et al. [283] which is recently applied
for cluster analysis by Liu et al. [284].
Gravitational Search Algorithm (GSA) – Rashedi [285,286] pro-posed the GSA following the principles of Newton law of
gravity which states that ‘Every particle in the universe attracts
every other particle with a force that is directly proportional to
the product of their masses and inversely proportional to the
square of the distance between them’. The algorithm is used for
cluster analysis by Hatamlou et al. [287]. Recently Yin et al.
[288] developed a hybrid algorithm based on K-harmonic
means and GSA.
2.3. Fitness functions for partitional clustering
The similarity function f ðÞ described in (2) plays a major role in
effectively partitioning the dataset. It represents a mathematical
function that quanties the goodness of a partition based on the
similarity between the patterns present in it. Various tness
functions used by the nature inspired metaheuristic algorithms
for partitional clustering are listed in Table 3.
2.4. Cluster validity indices
The cluster validity indices represent statistical functions used
for quantitative evaluation of the clusters derived from a dataset.The objective is to determine the importance of the disclosed
cluster structure produced by any clustering algorithm. In a recent
review article [289] Xu et al. compared the performance of eight
major validity indices used by swarm-intelligence-based cluster-
ing on synthetic and benchmark UCI datasets. Arbelaitz et al. [372]
have demonstrated the use of 30 cluster validity indices in 720
synthetic and 20 real datasets. The books by Gan et al. [95],
Berkhin [96] and Maulik et al. [322] present the validity indices
used by the evolutionary clustering algorithms. Some popular
validity indices like DB index, Dunn index, CS Measure and
Silhouette are also used as tness function by several researchers
(the details are enlisted in Table 3). Other validity indices used in
the bio-inspired clustering literature include CH Index [290,299],
I Index [112], Rand index [95], Jaccard coef cient [95], Folkes andMallows index [1], Hubert0s Γ statistic [95], SD Index [298], S-Dbwindex [295,296], root-mean-square standard deviation index
[295,296], RS index [297], PBM index [300] and SV index [301].
Gurrutxaga et al. [302] suggested a standard methodology to
evaluate internal cluster validity indices. Recently Saha and Ban-
dyopadhyay [303] have proposed connectivity based measures to
improve the performance of standard cluster validity indices used
by bio-inspired clustering techniques.
3. Multi-objective algorithms for exible clustering
Recent survey article by Zhou et al. [304] highlights the basic
principles, advancements and application of multi-objective algo-
rithms to several real world optimization problems. Basically these
algorithms are preferred over single objective counterparts as they
incorporate additional knowledge in terms of objective functions
to achieve optimal solution. In the last decade researchers have
developed many nature inspired multi-objective algorithms which
include non-dominated sorting GA (NSGA-II) [305,306], Pareto
envelope-based selection algorithm (PESA II) [325], Strength
Pareto Evolutionary Algorithm (SPEA) [326], and Voronoi Initia-
lised Evolutionary Nearest-Neighbour Algorithm (VIENNA) [327].
Along with these other major nature inspired multiobjective
algorithms are enlisted in Table 1. The recent book by Maulik
et al. [322] highlights the overview and applicability of these
multi-objective algorithms for partitional clustering.
3.1. Problem formulation
The partitional clustering problem can be formulated as a
multi-objective problem by simultaneously minimizing M objec-
tive function represented by
kAK Min
FðkÞ ¼ min½ f 1ðkÞ; f 2ðkÞ;…; f M ðkÞ ð9Þ
where K is the set of feasible clusters derived from dataset Z N D. In
multi-objective clustering, instead of achieving a single solution
(cluster partition achieved in single objective algorithm), a group
of optimal solutions are obtained (known as Pareto optimal) by
suitable combination of different objective functions. All the
Pareto optimal solutions are better from each other in the
form of some objective functions and therefore known as
S.J. Nanda, G. Panda / Swarm and Evolutionary Computation ∎ (∎∎∎∎) ∎∎∎–∎∎∎10
Please cite this article as: S.J. Nanda, G. Panda, A survey on nature inspired metaheuristic algorithms for partitional clustering, Swarmand Evolutionary Computation (2014), http://dx.doi.org/10.1016/j.swevo.2013.11.003i
http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003
8/16/2019 Nanda and Panda 2013 - A Survey on Nature Inspired Metaheuristic Algorithms for Partitional Clustering
11/18
non-dominated solutions [305]. The pictorial representation of
Pareto optimal solutions, with respect to the objective function is
known as Pareto optimal front [306].
3.2. Historical development in multi-objective algorithms for
partitional clustering
The survey paper by Bong and Rajeswari [323] reports that the
design, development and uses of multi-objective bio-inspiredalgorithms for clustering and classication problems have expo-
nentially increased from year 2006 to 2010. Research in the area of
bio-inspired multiobjective clustering has been strengthened after
the work on MOCK (Multi-objective clustering with automatic K
determination) by Handl and Knowles [330] published in 2007.
Prior to that Corne et al. developed pareto envelope-based selec-
tion algorithm (PESA) [324] and PESA-II [325] to solve partitional
clustering problem. Then the work by Handl and Knowles on
VIENNA (Voronoi Initialised Evolutionary Nearest-Neighbour Algo-
rithm) [327], multi-objective clustering with automatic determi-
nation of the number of clusters [328] and improvements in the
scalability [329] have drawn the attention of many evolutionary
computing researchers. These articles are considered to have
served as backbone for the development of MOCK [330].On the same year with MOCK [330], Bandyopadhyay et al. in
[331] reported multi-objective clustering based on NSGA-II [306]
and applied it for classication of remote sensing images. The
NSGA-II based multi-objective clustering has recently been applied
for MR brain image segmentation in [332]. Santosh et al. [333]
have proposed a multi ant-colonies based multi-objective cluster-
ing that can effectively group distributed data. Here each colony
works in parallel over the same dataset and simultaneous optimi-
zation of two objectives provides better solutions than those
achieved with individual objectives being separately optimized.
An immune-inspired algorithm to solve multiobjective cluster-
ing is initially proposed in [334] to classify the benchmark UCI
datasets UCI12. Then Ma et al. [335] developed the immunodomi-
nance and clonal selection inspired multiobjective clustering forclassifying handwritten digits. The immune multi-objective clus-
tering has been suitably applied for the SAR image segmentation
[336]. Recently Gou et al. [370] have reported development of
multi-elitist immune clonal quantum clustering algorithm.
An automatic kernel clustering using multi elitist PSO is
proposed by Das et al. in [337]. Paoli et al. [338] have formulated
the MOPSO based clustering for grouping hyperspectral images.
Recently the MOPSO has been applied for energy-ef cient cluster-
ing in mobile ad hoc networks [339].
Simulated annealing based multi-objective clustering algo-
rithm which uses symmetry distance is reported by Saha and
Bandyopadhyay [340,341]. A scatter tabu search algorithm is used
for multiobjective clustering problems in [342]. Suresh et al. [343]
have proposed a multi-objective differential evolution basedautomatic clustering for micro-array data analysis. The multi-
objective invasive weed optimization (MOIWO) has recently been
applied for cluster analysis by Liu et al. [284].
A clustering ensemble developed by Faceli et al. [344] deals
with generation of multiple partitions of the same data. Combining
these resulting partitions, an user can obtain a good data partition-
ing even though the original output clusters are not compact and
well separated. Ripon and Siddique have proposed an evolutionary
multi-objective tool for overlapping clusters detection.
3.3. Evaluation methods
Handl and Knowles [346] initially described the cluster validity
indices for multi-objective bio-inspired clustering. Then Brusco
and Steinley [347] reported the cross validation issues in multi-
objective clustering.
Recently the use of parametric and nonparametric statistical tests
has become popular among the evolutionary researchers. Usually
these tests are carried out to decide where one evolutionary
algorithm is considered better than another [348]. Therefore these
tests can effectively be applied to evaluate the performance of the
new multi-objective clustering algorithms. The parametric tests
described by Garcia et al. [349] are popular in which the authorshave selected 14 UCI datasets to compare the performance of ve
evolutionary algorithm used for classication purpose. They have
used Wilcoxon signed-ranks to evaluate the performance with
classication rate and Cohen0s kappa as accuracy measure. However
the parametric tests are based upon the assumptions of indepen-
dence, normality, and homoscedasticity which at times do not get
satised in multi-problem analysis. Under such situations the non-
parametric test is preferable. The papers by Derrac et al. [348] and
Garcia et al. [350] clearly highlight the signicance of nonparametric
test, which can perform two classes of analysis pairwise comparisons
and multiple comparisons. The pairwise comparisons include Sign
test, Wilcoxon test, Multiple sign test, and Friedman test. The multi-
ple comparisons consist of Friedman Aligned ranks, Quade test,
Contrast Estimation. The books [353,352] and statistical toolbox in
MATLAB [351] are helpful in implementing these statistical tests.
4. Real life application areas of nature inspired metaheuristics
based partitional clustering
The nature inspired partitional clustering algorithms have been
successfully applied to diversied areas of engineering and science.
Many researchers have employed the benchmark UCI datasets to
validate the performance of nature inspired clustering algorithms.
Some popular UCI datasets and its uses in the corresponding
algorithms are listed in Table 4. The major applications of the
nature inspired clustering literature and the corresponding authors
are shown in Table 5. Along with Table 5 some more applicationareas include character recognition [10,335], traveling salesman
problem [91], blind channel equalizer design [21], human action
classication [22,363], book clustering [32], texture segmentation
[210], tourism market segmentation [371], analysis of gene expres-
sion patterns [365], electrocardiogram processing [214], security
assessment in power systems [239], manufacturing cell design
[242], clustering of sensor nodes [362], identication of clusters for
accurate analysis of seismic catalogs [364].
5. Conclusion
This paper provides an up-to-date review of nature inspired
metaheuristic algorithms for partitional clustering. It is observedthat the traditional gradient based partitional algorithms are
computationally simpler but often provide inaccurate results as
the solution is trapped in the local minima. The nature inspired
metaheuristics explore the entire search space with the population
involved and ensure that optimal partition is achieved. Further
single objective algorithms provide one optimal solution where as
the multi-objective algorithms provide the exibility to select the
desired solution from a set of optimal solutions. The promising
solutions of automatic clustering are much helpful as they do not
need apriori information about the number of clusters present in
the dataset. It is important to note that although numerous
clustering algorithms have been published considering various
practical aspects, no single clustering algorithm has been shown to
dominate the rest for all application areas.
S.J. Nanda, G. Panda / Swarm and Evolutionary Computation ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 11
Please cite this article as: S.J. Nanda, G. Panda, A survey on nature inspired metaheuristic algorithms for partitional clustering, Swarmand Evolutionary Computation (2014), http://dx.doi.org/10.1016/j.swevo.2013.11.003i
http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003http://dx.doi.org/10.1016/j.swevo.2013.11.003
8/16/2019 Nanda and Panda 2013 - A Survey on Nature Inspired Metaheuristic Algorithms for Partitional Clustering
12/18
6. Future research issues
The eld of nature inspired partitional clustering is relatively
young and emerging with new concepts and applications. There
are many new research directions in this eld which need
investigations include:
In order to solve any partitional clustering problem the success of a particular nature inspired metaheuristic algorithm to achieve
optimal partition is dependent on its design environment (i.e
encoding scheme, operators, set of parameters, etc.). So for a given
complex problem the design choices should be theoretically
analyzed before the simulation and implementation.
Table 5
Real life application areas of nature inspired metaheuristic based partitional clustering.
Appl ica ti ons Popu lar researc h a rtic les b ased on n ature in sp ired pa rtitional c lu steri ng
Image segmentation GA – Feng et al. [130], PSO – Lee et al. [241], Abraham et al. [4], Zhang et al. [230], ACO – Ghosh et al. [190], DE – Das et al.
[2], Review – Jain et al. [10], NSGA II – Mukhopadhyay et al. [332], Bandyopadhyay et al. [331], MOCLONAL – Yang et al.
[336], Multiobj. Review – Bong and Rajeswari [323]
Image clustering GA – Bandyopadhyay et al. [113], DE – Das et al. [151,152], Omran et al. [156], PSO – Omran et al. [219,243],
NSGA II – Bandyopadhyay et al. [331]
Document clustering GA – Casillas et al. [115], Kuo and lin [129], PSO – Cui et al. [244], ACO – Yang et al. [194], Handl and Meyer [196],
DE – Abraham et al. [157], Review – Steinbach et al. [35], Andrews et al. [33], Jain et al. [34]
Web mining ACO – Labroche et al. [209], Abraham and Ramos [216], PSO – Alam et al. [245]
Text mining ACO – Handl and Meyer [196], Vizine et al. [218], SA - Chang [163]
Clustering in wireless sensor networks GA – Tan et al. [137], ABC – Karaboga et al. [251], Udgata et al. [252], PSO – Yu et al. [237], BFO – Gaba et al. [268], MOCSO
– Pradhan and Panda [20], Review – O. Younis et al. [17], M. Younis et al. [18], Kumarawadu et al. [19]
Clustering in mobile networks ACO – Merkle et al. [217], PSO – Ji et al. [238], Ali et al. [339], DE – Chakraborty et al. [158]
SA – W Jin et al. [164]
Gene expression data analysis GA – Lu et al. [109], Ma et al. [117], ACO – He and Hui [215], DE – Das et al. [2], PSO – Sun et al. [225], Du et al. [229],
Thangavel et al. [236], AIS –
Lie et al. [260], Review –
Jiang et al. [30], Lukashin et al. [31], Xu and Wunsch [91], Hruschkaet al. [3,119,120], Jain et al. [34], MODE – Suresh et al. [343]
Intrusion detection GA – Liu et al. [116], ACO – Ramos and Abraham [211], Tsang and Kwong [212], PSO – Lima et al. [246]
Computational nance Review – MacGregor et al. [24], Brabazon et al. [25], Amendola et al. [26], Nanda et al. [27]
Large datasets analysis GA – Franti et al. [124], Lucasius et al. [121], ACO – Chen et al. [213] Evolutionary algorithm NOCEA – Saras et al. [29]
Geological data analysis PSO – Cho [355], Review – Jain et al. [10,34], Zaliapin et al. [28],
Nanda et al. [364]
Table 4
Widely used UCI benchmark data sets for nature inspired metaheuristics based partitional clustering.
Datasets Creater Used in popular research articles for partitional clustering
Iris [150 4], Cl-3 R.A. Fisher GA – [100,114,91,128,134], DE – [16,155,153],
ACO – [184,198,193,202,207], BFO – [267],
PSO – [223,226,233,224,222,227], CSO – [271],
ABC – [247,250,248], Firey – [278], Frog – [180],
NSGA II – [345], MOAIS – [334], MOCK – [22], MODE – [343]
Wine [178 13], Cl-3 Forina et al. GA – [128], ACO – [201,189,193,202,207],
PSO – [223,224,233,226,227,231], DE – [16,155],
BFO – [267], AIS – [235], ABC – [247,250,248],
Firey – [278], GSA – [288], Frog – [180], NSGA II – [345],
MODE – [343], MOSA – [340,341], VIENNA – [327]
Glass [214 9], Cl-6 B. German GA – [4,128,134], ACO – [201,189,193,202,231],
PSO – [224,226,233,222,227], DE – [16,155],
BFO – [267], ABC – [250,248], Firey – [278], CSO – [271],
GSA – [288], NSGA II – [345]
Brest cancer [683 9], Cl-2 W.H. Wolberg GA – [4,134], ACO – [201,231], DE – [16,155],
O. Mangasarian PSO – [223,236,224,226,227], BFO – [267],
ABC – [250,248], Firey – [278], GSA – [288], MODE – [343],
MOAIS – [334], MOSA – [340,341], VIENNA – [327]
Thyroid [215 5], Cl-3 R. Quinlan ACO – [189,193], PSO – [226], ABC – [247,250],
Firey – [278], Frog