Systemic Functional Features in Stylistic Text Classification.

We propose that textual 'style' should be best de- fined as 'non-denotational meaning', i.e., those as- pects of a text's meaning that are mostly inde- pendent of what the text refers to in the world. To make this more concrete, we describe a lin- guistically well-motivated framework for compu- tational stylistic analysis based on Systemic Func- tional Linguistics. This theory views a text as a realisation of multiple overlapping choices within a network of related meanings, many of which re- late to non-denotational (traditionally 'pragmatic') aspects such as cohesion or interpersonal distance. Variations between relative frequencies of options within these systems corresponds to stylistic varia- tion. Though full parsing in SFL remains a difficult unsolved problem, we present a software architec- ture which allows for efficient modelling and ex- traction of SFL entities for use in stylistic analyses. In support, we show results for two applications: classifying texts as financial scams and classifying scientific articles by field.