Automatic text analysis: gradations of text types in web pages

Abstract

In this paper we present a first implementation of a linguistic expert system, which combines Bayesian inference and the knowledge of a domain expert, the linguist. The system is capable of automatically deriving the gradations of four text types -- descriptive/narrative, explicatory/informational, argumentative/persuasive, instructional -- from web pages. The evaluation of the system against human assessment shows that the system is effective and promising.