Image and Text Analysis using Multi-modal Embeddings
Abstract
While recent years witnessed an increasing use of computational text analysis in communications, the computational analysis of visual content remained a challenging task. Recent advancement in multi-modal embeddings offers promising solutions to conduct automated content analysis on images as well texts. Using an annotated dataset of green influencer posts on Instagram, this project presents and compares various inductive and deductive approaches to use pre-trained multi-modal embeddings for extracting theoretically relevant topics and frames from multi-modal content.
You may also be interested in the accompanying Masterclass for this event.
Speaker Biography
Justin Chun-ting Ho is a postdoctoral researcher at the Amsterdam School of Communication Research. Before Amsterdam, he worked at Academia Sinica in Taipei and Sciences Po in Paris. He holds a PhD in Sociology from the University of Edinburgh. His work focuses on nationalism, social media analysis, and computational methods.
Booking Information
Upon booking, attendees will receive an automated confirmation email which will contain a link to the online event page. This event will take place on Zoom. Attendees will receive a second email closer to the event booking which will also provide the link to the online event page. Please use the same email for registering with Eventbrite and when logging into Zoom.
Event Information
This is a webinar, which means that attendees’ microphones and cameras will be off. There will be opportunities for participants to turn their microphone on and ask questions during the discussion. This event will have live captions. Please inform us of any access requirements by emailing cdcs@ed.ac.uk. Further details about how CDCS uses your information obtained from booking onto our events can be found at in our Events Privacy Statement.