Visual cross-platform analysis: digital methods to research social media images

Pearce, Warren; Özkula, Suay Melisa; Greene, Amanda K.; Teeling, Lauren; Bansard, Jennifer S.; Omena, Janna Joceli; Rabello, Elaine Teixeira

doi:10.1080/1369118x.2018.1486871

Cited by 150 publications

(132 citation statements)

References 35 publications

Supporting

Mentioning

125

Contrasting

Unclassified

Order By: Relevance

“…Therefore while the overall progression informed my ideas of moving from a compositional to semiotic analysis within CVA, I rejected the quantitative steps. It should be noted that a similar mixed approach is frequently used across visual studies (Pearce et al ., ). Looking more specifically at studies of web images, I reviewed examples of thematic analysis (Duffy and Hund, ), content analysis with an emphasis on visual clustering (Delmestri et al ., ) and montage (Smith, ).…”

Section: Examining Pre‐existing Images: Exploring the Methodological Gapmentioning

confidence: 97%

“…Here I am not digging beneath platform algorithms but, as explored in depth by Pearce et al . (), note that researchers need to be aware of the implications of, and reflect fully on, platform choice. However, without significant funding and skilled resource unpacking the implications of platformisation is highly problematic.…”

Section: Cva Development: Research Journey and Organising Datamentioning

confidence: 99%

“…Critically there is now recognition that, across various internet media, a visual repertoire develops alongside a lexical vocabulary (Elliott and Robinson, ; Pearce et al ., ). In particular, Machin (.…”

Section: Introductionmentioning

confidence: 97%

See 2 more Smart Citations

Examining Web Images: A Combined Visual Analysis (CVA) Approach

Pritchard

2020

European Management Review

View full text Add to dashboard Cite

In this methodological paper I set out a framework for Combined Visual Analysis (CVA), bringing together compositional, reflexive and semiotic analysis. I explain how CVA was applied in a research project exploring the visual repertoire of human resource management (HRM). I describe each stage in detail, consider how research practice is instrumental in shaping research outcomes and reflexively explore the challenges encountered. The CVA framework provides a research protocol for those working with (in visual analytic terms) large numbers of preexisting images. It offers an approach that enables breadth and depth, while maintaining a qualitative focus on the images themselves.

show abstract

Section: Examining Pre‐existing Images: Exploring the Methodological Gapmentioning

confidence: 97%

Section: Cva Development: Research Journey and Organising Datamentioning

confidence: 99%

See 1 more Smart Citation

Examining Web Images: A Combined Visual Analysis (CVA) Approach

Pritchard

2020

European Management Review

View full text Add to dashboard Cite

show abstract

“…In a large-scale pilot study, O'Halloran et al (2016O'Halloran et al ( , 2018 used multimodal analysis visualisation to investigate where images of a particular journal were republished and how they were recontextualised. Similarly, Pearce et al (2018) developed various methods from network analysis to image sorting to identify similar images and their circulation in online communities.…”

Section: Circulation Of Images In Online Spacesmentioning

confidence: 99%

“…Both the number of Facebook communities to which each global platform provides images (measured through degree) and the number of images taken from each site (measured through strength) are higher than for Finnish sites and non-Finnish sites. Similarly, recent research on visual materials has suggested that studying these platforms is critical (Gibbs et al, 2015;Pearce et al 2018). Furthermore, the recent emergence of critical platform studies could prompt further investigation of the design and implementation of global platforms.…”

Section: Degree Strengthmentioning

confidence: 99%

Affective economy of national-populist images: Investigating national and transnational online networks through visual big data

Hokka

Nelimarkka

2019

New Media & Society

View full text Add to dashboard Cite

In our article, we investigate the affective economy of national-populist image circulation on Facebook. This is highly relevant, since social media has been an essential area for the spread of national-populist ideology. In our research, we analyse image circulation as affective practice, combining qualitative and quantitative methods. We use computational data analysis methods to examine visual big data: image fingerprints and reverse image search engines to track down the routes of thousands of circulated images as well as make discourse-historical analysis on the images that have gained most attention among supporters. Our research demonstrates that these existing tools allow social science research to make theory-solid approaches to understand the role of image circulation in creating and sustaining national and transnational networks on social media, and show how national-populist thinking is spread through images that catalyse and mobilise affects -fear, anger and resentment -thus creating an effective affective economy.

show abstract

Do you see what I see? Measuring the semantic differences in image‐recognition services' outputs

Berg,

Nelimarkka

2023

Asso for Info Science & Tech

View full text Add to dashboard Cite

As scholars increasingly undertake large‐scale analysis of visual materials, advanced computational tools show promise for informing that process. One technique in the toolbox is image recognition, made readily accessible via Google Vision AI, Microsoft Azure Computer Vision, and Amazon's Rekognition service. However, concerns about such issues as bias factors and low reliability have led to warnings against research employing it. A systematic study of cross‐service label agreement concretized such issues: using eight datasets, spanning professionally produced and user‐generated images, the work showed that image‐recognition services disagree on the most suitable labels for images. Beyond supporting caveats expressed in prior literature, the report articulates two mitigation strategies, both involving the use of multiple image‐recognition services: Highly explorative research could include all the labels, accepting noisier but less restrictive analysis output. Alternatively, scholars may employ word‐embedding‐based approaches to identify concepts that are similar enough for their purposes, then focus on those labels filtered in.

show abstract

Visual cross-platform analysis: digital methods to research social media images

Cited by 150 publications

References 35 publications

Examining Web Images: A Combined Visual Analysis (CVA) Approach

Examining Web Images: A Combined Visual Analysis (CVA) Approach

Affective economy of national-populist images: Investigating national and transnational online networks through visual big data

Do you see what I see? Measuring the semantic differences in image‐recognition services' outputs

Contact Info

Product

Resources

About