Skip to main content

Advertisement

Table 6 A comparison between our method (EPSO) and previous PSO-based methods

From: An enhancement of binary particle swarm optimization for gene selection in classifying cancer classes

Data Method EPSO [This work] IBPSO[6] PSOTS[3] PSOGA[7] TS-BPSO[8] BPSO-CGA[9]
  Evaluation       
11_Tumors Average #Acc (%) 95.40 - - - - -
Best #Acc (%) 96.55 93.10 - - 97.35 -
Average #Genes 237.70 - - - - -
Best #Genes 243 2948 - - 3206 -
9_Tumors Average #Acc (%) 75 - - - - -
Best #Acc (%) 76.67 78.33 - - 81.63 -
Average #Genes 247.10 - - - - -
Best #Genes 251 1280 - - 2941 -
Brain_Tumor1 Average #Acc (%) 92.11 - - - -  
Best #Acc (%) 93.33 94.44 - - 95.89 91.4
Average #Genes 7.5 - - - - -
Best #Genes 8 754 - - 2913 456
Brain_Tumor2 Average #Acc (%) 92.4 - - - - -
Best #Acc (%) 94 94.00 - - 92.65 -
Average #Genes 6.0 - - - - -
Best #Genes 4 1197 - - 5086 -
Leukemia1 Average #Acc (%) 100 - 98.61 95.10 - -
Best #Acc (%) 100 100 - - 100 100
Average #Genes 3.2 - 7 21 - -
Best #Genes 2 1034 - - 2577 300
Leukemia2 Average #Acc (%) 100 - - - - -
Best #Acc (%) 100 100 - - 100 -
Average #Genes 6.8 - - - - -
Best #Genes 4 1292 - - 5609 -
Lung_Cancer Average #Acc (%) 95.67 - - - - -
Best #Acc (%) 96.06 96.55 - - 99.52 -
Average #Genes 8.3 - - - - -
Best #Genes 7 1897 - - 6958 -
SRBCT Average #Acc (%) 99.64 - - - - -
Best #Acc (%) 100 100 - - 100.00 100
Average #Genes 14.90 - - - - -
Best #Genes 7 431 - - 1084 880
Prostate_Tumor Average #Acc (%) 97.84 - - - - -
Best #Acc (%) 99.02 92.16 - - 95.45 93.7
Average #Genes 6.6 - - - - -
Best #Genes 5 1294 - - 5320 795
DLBCL Average #Acc (%) 100 - - - - -
Best #Acc (%) 100 100 - - 100.00 -
Average #Genes 4.7 - - - - -
  Best #Genes 3 1042 - - 2671 -
  1. Note: The best result of each evaluation is written in a bold style. The best result of evaluations could not be compared and stated if the results of evaluations have not been reported in the previous works. The best method of each data set is shown in the shaded cells. The best method is selected based on the highest number of the best results for all evaluations. #Acc and S.D. denote the classification accuracy and the standard deviation, respectively, whereas #Genes and #Ave represent the number of selected genes and an average, respectively. #Time stands for running time in the hour unit. ‘-‘ means that a result is not reported in the previous related work.
  2. IBPSO = An improved binary PSO. PSOTS = A hybrid of PSO and tabu search.
  3. PSOGA = A hybrid of PSO and GA. TS-BPSO = A combination of tabu search and BPSO.
  4. BPSO-CGA = A hybrid of BPSO and a combat genetic algorithm.