Research Article

Improved Instance Selection Methods for Support Vector Machine Speed Optimization

Table 4

Result for FFA_IS.

Total emailNFFNIAPA (%)GBFP (%)FN (%) (%)Pr (%)FM (%) (s)

4000510099.251000.294.0096.0097.9896.8323.54
4000515099.551000.202.2097.8098.6798.1139.07
4000520099.551000.112.8097.2099.2098.0862.24
4000525099.601000.172.0098.0098.8398.3279.12
4000530099.651000.112.0098.0099.2298.49109.70
40001010099.381000.173.8096.2098.8797.2828.57
40001015099.551000.261.8098.2098.2998.1345.94
40001020099.631000.231.4098.6098.4698.4684.30
40001025099.681000.092.0098.0099.4298.59117.98
40001030099.601000.172.0098.0098.8398.30213.17

Key: NFF: number of fireflies, NI: number of instances, GB: global best, APA: average prediction accuracy, FP: false positive, FN: false negative, : recall, Pr: precision, FM: -measure, and : time.