Track:
Contents
Downloads:
Abstract:
Typically, high volume text processing has been based on representations of the texts such as Parts of Speech (POS), parsed formalizations or bag of words. This paper proposes an alternative representation of the text based on linguistic analysis. Drawing on systemic functional theory and its application to stylistics, we will develop a text phase portrait' comprised of the patterns of systemic features that are foregrounded in each text. We will only focus on computationally tractable features to ensure that textual phase portraits are easily identifiable by machine learner technology, a method that has proved successful in previous similar research.