Using Machine Learning to investigate censorship.
Abstract
Machine Learning and Natural Language Processing emphasizes learning style and predictive power in which predictive text is generated through feeding in a large amount of training data. But moving beyond accurate prediction as the end goal, how could we raise different kinds of questions through working with big data and machine learning processes?
In this lecture, I will present the work-in-progress called "Unerasable Characters”, which utilises a character-based machine learning algorithm (a recurrent neural network) to take input from the censored text of the Chinese Social Media Platform Weibo as the source and output generative text to think through questions related to erasure, authorship, poetic text and cultural implications of technology. The talk suggests machine learning as a material-discursive object, allowing researchers and practitioners to zoom-in on specific material instances and actions, as well as reflecting on wider conceptual and cultural issues.
bio
Winnie Soon is an artist-researcher whose works intersect media/computational art, software studies, cultural studies and code practice. Her research and practice is dedicated to the cultural implications of technologies where computational processes are increasingly manifesting our programmed experiences. Her works deal with topics around computational culture, in particular regard to automated censorship, data circulation, real-time processing/liveness, invisible infrastructure and the culture of code practice.
Winnie’s projects have been presented and exhibited internationally at museums, art festivals, libraries, universities and conferences, including but not limited to ZKM, RMIT Gallery, The Photographers’ Gallery, Transmediale, Electronic Literature Festival, ISEA, Stuttgarter Filmwinter, WRO Media Art Biennale, Roskilde Library, Image Galleri, Si Shang Art Museum, Pulse Art + Technology Festival, FutureEverything Art Exhibition, Ars Electronica, The Wrong – New Digital Art Biennale, Hong Kong Microwave International Media Arts Festival, and among others. Her current research focuses on critical technical practice, working on two books titled “Aesthetic Programming: A Handbook of Software Studies” (with Geoff Cox) and “Fix My Code” (with Cornelia Sollfrank). Recent contributions to publications include “Execution” in Posthuman Glossary (with Critical Software Thing), ‘Executing Microtemporality’ in Executing Practices and “API practices and paradigms” in First Monday (with Eric Snodgrass).
Currently, she is Assistant Professor in the Department of Digital Design at Aarhus University. More info: http://www.siusoon.net
Dates & times
| Date |
Time |
Add to calendar |
| 13 Nov 2019 |
5:00pm - 7:00pm |
|
Accessibility
If you are attending an event and need the College to help with any mobility requirements you may have, please contact the event organiser in advance to ensure we can accommodate your needs.