Thoughts sent to faculty about summer time independent study.

Part of unrevdb

---------- Forwarded message ----------
Date: Wed, 3 Apr 2002 20:49:27 -0500 (EST)
From: cdent@burningchrome.com
To: ralf
Subject: Independent study possibilities


Hello,

This is Chris Dent. As you may know I've been working with
Kathryn La Barre on her work to do some analysis of the unrev-ii
mail archive. That work has taken on a life of its own and is
going to extend beyond the end of the semester. We've made some
contacts both at the Bootstrap Alliance (Engelbart's foundation)
and the PORT project (a project headed by Mark Keeler that is
working on tools for accessing archives of Peirce manuscripts and
information (annotations, corrections, etc.) about those manuscripts).

There is a great deal to do here. My initial involvement was as a
helper, I believe I was paid out of some of your own funds. Thank
you very much for that.

My involvement has since grown to where I am now a
co-investigator on the project. Since it is likely that the
project will continue I'd like to know what the possibilities are
for getting credit over the summer for the work. Kathryn
suggested you as a good person to ask.

I'll start with this simple question and hopefully we can proceed
from there: What do I need to provide to make this a valid
pursuit as an independent study?

What I will be doing is investigating methods for automating the
generation or identification of facets in the dataset. Thus far
we've had some success using LSA to cluster documents into
possible sections of aboutness. From there, on the advice of Uta,
we hope to be able to label the clusters (by human hand) and then
turn around and go the other direction: map documents to the
discovered topics.  This will allow for a somewhat automatic
incorporation of new documents.

LSA has not really proven itself yet (in this context), so we'd
like to investigate other clustering tools as well, and tools for
facilitating human identification of "aboutness". Tonight I'm
planning to start on a web interface that will be a combintation
of a search interface to the archive as well as a collaborative
tool for labelling each message with descriptive tags that will
be generated by the "team" using the interface.

These are but a subset of a Kathryn's goals; my particular focus.

Thanks for your comments.