Hi, everyone! I would love to meet my fellow AI-in-libraries nerds, so how about we introduce ourselves and our interests in AI?
I'm Andromeda Yelton, a librarian/software developer in greater Boston (and former LITA president), as well as adjunct faculty at SJSU -- where I teach
a class on machine learning :). I'm really interested in using machine learning to write love letters to library data -- what kinds of storytelling, exploration, research, and delight can we open up with these tools? How do we do this in an intellectually rigorous and socially responsible way?
I'm particularly psyched about my development of
HAMLET (How About Machine Learning Enhancing Theses), a neural net I trained on ~43,000 graduate theses in order to find out what kinds of discovery interfaces I could build for a collection that doesn't have traditional subject headers. I had some fun
visualizing that recently. I've also recently learned about neural style transfer, enabling me to use DPLA images to make
Christmas cards from the void. Lately I've been exploring means for improving face recognition in archival photographs.
How about y'all?
------------------------------
Andromeda Yelton
Humanistic Machine Learning for Library Data
Lecturer, San José State University iSchool
https://andromedayelton.com@ThatAndromeda
------------------------------