Glitter: Visualizing Lexical Surprisal for Readability in Administrative Texts

TL;DR

This paper introduces Glitter, a visualization tool that estimates text readability by measuring lexical surprisal and information entropy using multiple language models, aiming to improve administrative texts' clarity.

Contribution

The paper presents a novel visualization framework for estimating text readability through lexical surprisal and entropy, specifically targeting bureaucratic language.

Findings

Effective visualization of lexical surprisal for readability analysis

Potential to enhance clarity of administrative texts

Open-source tool available for practical use

Abstract

This work investigates how measuring information entropy of text can be used to estimate its readability. We propose a visualization framework that can be used to approximate information entropy of text using multiple language models and visualize the result. The end goal is to use this method to estimate and improve readability and clarity of administrative or bureaucratic texts. Our toolset is available as a libre software on https://github.com/ufal/Glitter.