Images in Social Media Research: Digital Tools and Methodological Challenges (Forum/Workshop)
Images in Social Media Research: Digital Tools and Methodological Challenges
Online-Workshop, 10th February 2023, 9.00-16.30 (CET)
From memes on Twitter and Reddit to Instagram posts and TikTok videos: Images are taking on increasingly important roles in social media communication. While social media research can draw on established tools in the field of text analysis, researching visual content still presents particular methodological challenges. How can images be systematically searched and scraped from the web? What qualitative and quantitative possibilities are there for structuring, visualising, and evaluating big image corpora and their metadata? How can a sustainable and critical approach to sensitive images look like?
This one-day workshop brought together image researchers from different disciplines. On a very hands-on level, we wanted to explore digital tools that might offer pragmatic solutions for researching images on the web. Furthermore, the academic (re)use of images from social media raises particular ethical and legal issues. Digital media researchers, legal scholars and art historians provided insight into their method designs and opened up critical perspectives on how to deal with visual content online.
Over the course of the three sessions, we had more than 500 visits, around 300 participants joined the webinar. This means that not only the majority of European core countries were represented, but also numerous students and researchers from Australia, Egypt, Ghana, Hong Kong, India, Israel, Japan, Canada, Colombia, Korea, Lebanon, Mexico, Pakistan, Puerto Rico, Russia, Singapore, South Africa, Turkey and Ukraine. Social media, the tools and methods of their exploration seem to be a literal cross-cutting topic.
The workshop was accompanied by numerous comments and questions, almost 80 were collected in the Q&A alone, some of which were answered live. Some answers as well as further information were collected on an etherpad during the course of the workshop. Not only platforms such as Facebook, Twitter, Reddit, YouTube, Instagram, TikTok and co were covered. The questions also targeted Telegram, Flickr, Pinterest and Steam. The pragmatic use of tools, but also differences between fair use (USA), copyright and CC licences as well as ethical questions around the publication of structured raw and sensitive data were addressed.
We do not share any recordings, but please find some of the presentations as well as the archived etherpad with further links and recommendations below.
Programme (PDF / archived etherpad)
9.00 Verena Straub & Christoph Eggersglüß: Welcome and Introduction
- Bernhard Rieder: Analysing YouTube through data extraction – Introducing the YouTube data tools
- Stijn Peeters: Quali-quantitative internet research with 4CAT (slides)
- Jason Chao: Enriching image data with AI using Memespector-GUI
- Jens-Martin Loebel: Exploring Yenda and HyperImage – A scientific approach to networked digital image annotation and hypermedia publication
- Sabine Niederer: Visual methodologies for networked images (slides)
- Elena Pilipets: Deep TikTok – Three methods for tracing video memes (slides)
- Janna Joceli Omena & Richard Rogers: Analysing image collections with the computer vision network approach (slides)
- Lev Manovich: Protests, cities, selfies – How we visualise millions of social media images
15.00-16.30 Ethical and Legal Issues
- Grischka Petri: Private, social, public? Images on social media and some of their legal conditions
- Tanja-Bianca Schmidt: How can violence be appropriately addressed? Some thoughts on images and their ethical dilemma (slides)
- Evelyn Runge: Wandering images across platforms. Tracking alterations – An exploration
The workshop is organised by the research project Image Protests on Social Media at TU Dresden, funded by the German Research Foundation (DFG), in collaboration with NFDI4Culture, Consortium for Research Data on Material and Immaterial Cultural Heritage within the Nationale Forschungsdateninfrastruktur (NFDI).
Contact: Verena Straub, verena.straub (at) tu-dresden.de & Christoph Eggersglüß, christoph.eggersgluess (at) uni-marburg.de