DISCERN story processing demo

DISCERN is a large, modular neural network system for reading, paraphrasing and answering questions about stereotypical (script-based) stories. The main reference is

Miikkulainen, R. (1993). Subsymbolic Natural Language Processing: An Integrated Model of Scripts, Lexicon, and Memory. Cambridge, MA: MIT Press.

A precis of this book and a short summary paper are available through the Web, as is the source code for the entire DISCERN system.

This demo shows what DISCERN looks like in action. It displays the network activations graphically as the system is reading stories and answering question about them and prints out verbose log output. It is set up to process the example in the Subsymbolic NLP book; all you need to do is to click on the "Run" button a couple of times to walk through the example. For more interesting interactions with the program (such as commands and questions from the keyboard, examining the memory with the pointing device), see the user instructions.

For the demo to work, you must be running X11 (not Windows, Macintosh, or NeXTStep), and you must not have a firewall preventing all remote access to your screen. The demo program runs on net.cs.utexas.edu, with the display over the internet on your X11 screen. Send comments, bug reports etc. to discern@cs.utexas.edu.

Click here to proceed with the demo.

Back to UTCS Neural Networks home page
Last update: 1.8 2003/01/18 00:09:51 hongming