Data Mining

There’s an interesting article in the recent D-Lib Magazine: “From Babel to Knowledge: Data Mining Large Digital Collections,” by Daniel J. Cohen. He talks about the task of refining search tools in a couple of specific ways so that a researcher can extract the kind of document needed from the welter of uncatalogued documents on the internet (or in offline collections). It’s not hard to see the potential for getting better access to law-related documents or otherwise making better use of full-text-indexed law related databases.

You might take a particular look at two of the tools his research has led to, Syllabus Finder, which, as the name suggests, throws up academics’ syllabi on topics searched for, and H-Bot, a natural language search engine that can answer (some) questions about history.

[via this month’s Current Cites]