Understanding scholarly Natural Language Processing system diagrams through application of the Richards-Engelhardt framework

TL;DR

This paper applies the Richards-Engelhardt framework to analyze scholarly NLP system diagrams, highlighting the need for new visual encoding principles and vocabulary to better understand and describe complex system representations.

Contribution

It introduces the addition of 'Grouping by Object' and 'Emphasising' as new visual encoding principles for better diagram analysis.

Findings

Framework effectively reflects diagram complexity

Vocabulary needed for multiple-codings and semiotic variability

Proposes new visual encoding principles

Abstract

We utilise Richards-Engelhardt framework as a tool for understanding Natural Language Processing systems diagrams. Through four examples from scholarly proceedings, we find that the application of the framework to this ecological and complex domain is effective for reflecting on these diagrams. We argue for vocabulary to describe multiple-codings, semiotic variability, and inconsistency or misuse of visual encoding principles in diagrams. Further, for application to scholarly Natural Language Processing systems, and perhaps systems diagrams more broadly, we propose the addition of "Grouping by Object" as a new visual encoding principle, and "Emphasising" as a new visual encoding type.