Sensitivity and Specificity are ratios to describe how accurately a binary classifier labels each case, both the ones it should flag and the ones it should leave alone. Sensitivity is the chance the test flags a genuine positive. Specificity is the chance the test correctly returns negative for a genuine negative.