Review Article

Cat Swarm Optimization Algorithm: A Survey and Performance Evaluation

Table 1

Summary of the modified versions of the CSO algorithm.

Comparison ofWithTesting fieldPerformanceReference

CSO (original)PSO and weighted-PSOSix test functionsBetter[4, 5]
BCSOGA, BPSO, and NBPSOFour test functions (sphere, Rastrigin, Ackley, and Rosenbrock)Better[9]
MOCSONSGA-IICooperative spectrum sensing in cognitive radioBetter[10]
PCSOCSO and weighted-PSOThree test functions (Rosenbrock, Rastrigrin, and Griewank)Better when the number of iteration is fewer and the population size is small[11]
CSO clusteringK-means and PSO clusteringFour different clustering datasets (Iris, Soybean, Glass, and Balance Scale)More accurate but slower.[12]
EPCSOPCSO, PSO-LDIW, PSO-CREV, GCPSO, MPSO-TVAC, CPSO-H6, PSO-DVMFive test functions and aircraft schedule recovery problemBetter[13]
AICSOCSOThree test functions (Rastrigrin, Griewank, and Ackley)Better[14]
ADCSOCSOSix test functions (Rastrigrin, Griewank, Ackley, axis parallel, Trid10, and Zakharov)Better except for Griewank test function.[15]
Enhanced HCSOPSOMotion estimation block-matchingBetter[16, 17]
ICSOPSOMotion estimation block-matchingBetter[17]
OL-ICSOK-median, PSO, CSO, and ICSOART1, ART2, Iris, CMC, Cancer, and Wine datasetsBetter[18]
CQCSOQCSO, CSO, PSO, and CPSOFive test functions (Schaffer, Shubert, Griewank, Rastrigrin, and Rosenbrock) and multipeak maximum power point tracking for a photovoltaic array under complex conditionsBetter[19]
ICSOCSO and PSOThe 69-bus test distribution systemBetter[20]
ICSOCSO, BCSO, AICSO, and EPCSOTwelve test functions (sphere, Rosenbrock, Rastrigin, Griewank, Ackley, Step, Powell, Schwefel, Schaffer, Zakharov’s, Michalewicz, quartic) and five real-life clustering problems (Iris, Cancer, CMC, Wine, and Glass)Better[21]
Hybrid PCSOABCPCSO and ABCFive test functionsBetter[22]
CSO-GA-PSOSVMCSO + SVM (CSOSVM)66 feature points from each face of CK + (Cohn Kanade) datasetBetter[23]
Hybrid CSO-based algorithmGA, EA, SA, PSO, and AFSSchool timetabling test instancesBetter[24]
Hybrid CSO-GA-SASLPA and CFinderSeven datasets (Karate, Dolphin, Polbooks, Football, Net-Science, Power, Indian Railway)Better[25]
MCSOCSONine datasets from UCIBetter[26]
MCSOCSOEight datasetBetter[27]
NMCSOCSO, PSOSixteen benchmark functionsBetter[28]
ICSOCSOTen datasets from UCIBetter[29]
cCSODE, PSO, CSO47 benchmark functionsBetter[30]
BBCSOBinary particle swarm optimization (BPSO), binary genetic algorithm (BGA), binary CSO0/1 Knapsack optimization problemBetter[31]
CSO-CSN/AVRP instances from http://neo.lcc.uma.es/vrp/N/A[32]