Research Article

Fuzzy Rules for Ant Based Clustering Algorithm

Table 4

Mean, standard deviation, mode, min, and max values of error rate achieved by each clustering algorithm (-means, -medoid, FCM, and ASClass) over 20 trials on dataset presented in Table 2.

DatasetsAlgorithmMeanStdModeMinMax

Art1-means0.8060.1800.9470.52250.9950
-medoid0.7730.1580.7600.51500.9750
FCM0.5550.5550.55500.5550
ASClass0.47100.4710.47130.4713

Art2-means0.5480.4900.9800.0200.980
-medoid0.4040.4820.0200.0200.980
FCM0.9800.9800.9800.980
ASClass0.03900.0390.0390.039

Art3-means0.7490.2270.5850.2220.989
-medoid0.8100.1410.5490.5490.995
FCM0.64400.6440.6440.644
ASClass0.5070.5080.5060.508

Art4-means0.4500.510001
-medoid0.5720.494101
FCM0.9852.278 × 10−160.9850.9850.985
ASClass0.01500.0150.0150.015

Art5-means0.9080.0650.8730.8020.990
-medoid0.8830.1330.9850.4550.996
FCM0.9052.278 × 10160.9050.9050.905
ASClass0.476600.4760.4760.476

Art6-means0.6410.2730.75001
-medoid0.8270.17610.4271
FCM0.75000.7500.7500.750
ASClass0.0890.0020.0880.0850.095

Iris-means0.5760.3360.4800.1131
-medoid0.7100.2290.1260.1261
FCM0.7462.278 × 10−160.7460.7460.746
ASClass0.2575.695 × 10−170.2570.2570.257

Thyroid-means0.7740.1660.9220.2390.922
-medoid0.5940.2610.6660.1280.952
FCM0.8472.278 × 10−160.8470.8470.847
ASClass0.47600.4760.4760.476

Pima-means0.4610.1250.3710.3710.628
-medoid0.5040.1160.6160.3250.641
FCM0.6501.139 × 10−160.6500.6500.650
ASClass0.5430.0080.5380.5300.565