A Diagnostic Model of Breast Cancer Based on Digital Mammogram Images Using Machine Learning Techniques

Table 4

The training and testing ratios distribution of samples from the MIAS database.

The class type	Number of samples	Training set (around 70% of samples)	Testing set (30% of samples)
Normal (no cancer)	208	157	51
Abnormal (malignant)	49	37	12
Abnormal (benign)	67	48	19
Total number	324	242	82