The Fewer Splits are Better: Deconstructing Readability in Sentence Splitting

TL;DR

This paper investigates whether splitting complex sentences into two or three parts affects readability, using Bayesian modeling and crowd-sourced evaluations, and finds that splitting into two parts generally improves understanding more than into three.

Contribution

It introduces a Bayesian framework to analyze sentence splitting effects on readability and provides empirical evidence favoring fewer splits for better comprehension.

Findings

Splitting into two parts improves readability more than splitting into three.

Bayesian modeling quantifies the impact of split number on comprehension.

Empirical data from crowd-sourced evaluations supports the preference for fewer splits.

Abstract

In this work, we focus on sentence splitting, a subfield of text simplification, motivated largely by an unproven idea that if you divide a sentence in pieces, it should become easier to understand. Our primary goal in this paper is to find out whether this is true. In particular, we ask, does it matter whether we break a sentence into two or three? We report on our findings based on Amazon Mechanical Turk. More specifically, we introduce a Bayesian modeling framework to further investigate to what degree a particular way of splitting the complex sentence affects readability, along with a number of other parameters adopted from diverse perspectives, including clinical linguistics, and cognitive linguistics. The Bayesian modeling experiment provides clear evidence that bisecting the sentence leads to enhanced readability to a degree greater than what we create by trisection.