Practical Skills of Business English Correspondence Writing Based on Data Mining Algorithm

Table 3

Sample statistics of training set and test set.

	Training set	Test set
Total number of labeled samples	2405	2145
Number of samples in category I	535	520
Category I sample proportion	0.22	0.24
Number of samples in category II	415	365
Category II sample proportion	0.16	0.16
Number of samples in category III	920	795
Category III sample proportion	0.32	0.36
Number of samples in category IV	530	466
Category IV sample proportion	0.22	0.21