Visference: Improving scientific papers representation using topic model analysis and web interface interactions.

See DEMO SITE

Get poster

Github CODE

This is a representation of 282 accepted papers from the Journal of Machine Learning Research (JMLR) Workshop and Conference Proceedings Volume 28: Proceedings of The 30th International Conference on Machine Learning (ICML).

Visference was developed with Dr Wray Buntine and Dr Mark Reid in 2013

 

This prototype is visually conservative in the sense that it tries to not scare usual visitors with a flashy visualization but incorporate known interactions, like table sorting. The tool, Visference, has evaluated positively by users and experts.

A poster and the demo has been published and presented in NICTA conferences and retreats from 2013-15.