Home Linguistics & Semiotics Chapter 7. Constructing corpora from images and text
Chapter
Licensed
Unlicensed Requires Authentication

Chapter 7. Constructing corpora from images and text

An introduction to Visual Constituent Analysis
  • Alex Christiansen , William Dance and Alexander Wild
View more publications by John Benjamins Publishing Company
Corpus Approaches to Social Media
This chapter is in the book Corpus Approaches to Social Media

Abstract

Visual analysis represents a significant oversight in the corpus literature, and possibly one that may lead to unintended omissions, particularly when analysing social media. In this chapter we introduce Visual Constituent Analysis (VCA), a method of multimodal corpus construction that allows researchers to construct and analyse visual aspects of online media in large-scale corpora. The chapter addresses the shortcomings of a purely textual approach to discourse analysis when dealing with social media texts and offers a solution using computer ‘Vision’-based image annotation (in our case Google Cloud Vision). Finally, we demonstrate how our approach can be used to analyse a sample of 150,000 micro-blog posts from Twitter and show the difference in level of user interaction with combined image/texts over language-only social media texts.

Abstract

Visual analysis represents a significant oversight in the corpus literature, and possibly one that may lead to unintended omissions, particularly when analysing social media. In this chapter we introduce Visual Constituent Analysis (VCA), a method of multimodal corpus construction that allows researchers to construct and analyse visual aspects of online media in large-scale corpora. The chapter addresses the shortcomings of a purely textual approach to discourse analysis when dealing with social media texts and offers a solution using computer ‘Vision’-based image annotation (in our case Google Cloud Vision). Finally, we demonstrate how our approach can be used to analyse a sample of 150,000 micro-blog posts from Twitter and show the difference in level of user interaction with combined image/texts over language-only social media texts.

Downloaded on 14.9.2025 from https://www.degruyterbrill.com/document/doi/10.1075/scl.98.07chr/html?lang=en
Scroll to top button