Table 2.

Statistical text mining modeling parameters.

Algorithm	Parameters
Decision trees
Term weighting	GR, LOR, X2
Top n terms	10, 25, 50, 100, 250, 500, 1000, All
Split criterion	GI, GR
k-Nearest neighbor
Term weighting	GR, LOR, X2
Top n terms	25, 50, 100, 250, 500, 1000, All
k	1, 2, 5, 10
Support vector machines
Term weighting	GR, LOR, X2
Top n terms	0, 25, 50, 100, 250, 500, 1000, All
SVD dimensions	0, 25, 50, 100, 250

Abbreviations: GI, Gini index; GR, gain ratio; LOR, log odds ratio; X², Chi-square.