Research Article

Uncertainty Measurement and Attribute Reduction Algorithm Based on Kernel Similarity Rough Set Model

Table 3

Comparison of SVM classification accuracy of attribute reduction results

Data setsOriginal dataThe comparison Algorithm 1The comparison Algorithm 2The comparison Algorithm 3The algorithm proposed in this paper

wine0.9250 ± 0.00270.9493 ± 0.00640.9526 ± 0.00570.9372 ± 0.00270.9657±0.0019
sonar0.8066 ± 0.00900.8550 ± 0.01720.8425 ± 0.01450.8237 ± 0.01260.8564±0.0132
iono0.8109 ± 0.00430.8432 ± 0.00430.8521 ± 0.00480.8329 ± 0.00620.8570±0.0057
wdbc0.9251 ± 0.00220.9572 ± 0.00290.9735±0.00450.9356 ± 0.00180.9697 ± 0.0023
biodeg0.8250 ± 0.00270.8383 ± 0.00490.8551±0.00330.8290 ± 0.00270.8478 ± 0.0036
messidor0.8458 ± 0.00610.8557 ± 0.00670.8608 ± 0.00580.8593 ± 0.00690.8713±0.0062
winequality-red0.7671 ± 0.00740.8129±0.00810.7864 ± 0.00620.7781 ± 0.00580.7932 ± 0.0051
winequality-white0.7484 ± 0.00170.8364 ± 0.00770.8242 ± 0.00420.7992 ± 0.00650.8425±0.0047
magic0.8266 ± 0.00110.8596 ± 0.00230.8691 ± 0.00310.8431 ± 0.00260.8793±0.0018
average0.8311 ± 0.00410.8675 ± 0.00670.8684 ± 0.00570.8486 ± 0.00530.8758±0.0049

The best experimental results are highlighted in bold.