Research Article

Association Rule-Based Feature Mining for Automated Fault Diagnosis of Rolling Bearing

Table 3

The indicator of the RMS-based criteria based on different discretization methods.

ā€‰fBPFOfBSFfBPFIfxMVxSDxRMSxRAxCFxSxKxKC

Equal probability0.000.000.000.001.000.000.000.000.000.000.000.00
0.001.002.060.003.092.782.782.834.230.000.000.00
4.784.963.995.991.902.842.844.692.856.395.554.53
4.822.811.002.004.492.322.322.781.943.462.002.25
0.003.411.961.003.990.000.000.003.451.000.002.60

Equal density1.001.001.001.513.221.001.001.001.003.681.002.13
2.132.442.853.562.511.901.902.001.003.511.512.61
2.612.752.753.614.522.512.512.612.512.272.851.90
1.714.293.614.613.912.272.272.752.892.471.512.13
1.652.751.814.374.523.613.613.382.513.082.401.62

Equal width7.535.792.322.580.001.941.941.652.212.712.812.92
1.683.650.870.500.001.791.791.941.330.340.001.83
0.002.531.000.881.521.161.160.921.730.001.001.00
0.000.980.850.002.170.000.000.000.000.000.721.28
0.000.000.000.002.280.000.000.000.990.000.920.00