A Data-Oriented Model of Literary Language

TL;DR

This paper introduces a novel data-oriented model that predicts the literary quality of texts using lexical and syntactic features, achieving high accuracy in distinguishing levels of literaryness.

Contribution

It is the first model to differentiate degrees of literaryness in novels using a combination of mined syntactic fragments and hand-crafted features.

Findings

Explains 76.0% of variation in literary ratings.

Outperforms standard bigram baseline.

Utilizes rich syntactic features for literary assessment.

Abstract

We consider the task of predicting how literary a text is, with a gold standard from human ratings. Aside from a standard bigram baseline, we apply rich syntactic tree fragments, mined from the training set, and a series of hand-picked features. Our model is the first to distinguish degrees of highly and less literary novels using a variety of lexical and syntactic features, and explains 76.0 % of the variation in literary ratings.