I'm still a week or so away from finishing the diss, but I ran across something that really impressed me, and thought I'd pass it along. Also, I just completed my introductory chapter, and may be feeling a bit cocky.
The Department of Energy's Pacific Northwest National Laboratory has an Information Visualization project that is truly fascinating. The problem is that for many, many fields there is now more digital information available than a single human can meaningfully process. Visually representing patterns found in text, images, and audio filters out "insignificant" variation, allowing us to deal with greater data volumes.
I recommend checking out their published papers.
Well, back to it. See you in a couple weeks.
Tags: text mining, visualization