Word frequency–rank relationship in tagged texts

We analyze the frequency–rank relationship in sub-vocabularies corresponding to three different grammatical classes (nouns, verbs, and others) a collection of literary works English, whose words have been automatically tagged according their role. Comparing with null hypothesis which assumes that belonging each class are uniformly distributed across frequency–ranked vocabulary whole work, we disclose statistically significant differences between classes. This results point fact relationships may reflect linguistic features associated function.