blog: The potential of text-mining tools for historical scholarship
The Center for History and New Media at George Mason University will undertake a major two-year study of the potential of text-mining tools for historical (and by extension, humanities) scholarship. Read more about this exciting project on Dan Cohen's blog.
As some of you will know, the Methods Network has supported two workshops on text mining, both of which had a special focus on historical applications of this method:
http://www.methodsnetwork.ac.uk/activities/act25.html
http://www.methodsnetwork.ac.uk/activities/act6.html
One thing that was mentioned during these events (and that partly prompted us to fund the second one) is that historians overall are much more reluctant to use text mining than you would expect - after all, text is still the major source for most historians. As it happens, I was just discussing the second workshop with Zoe Bliss the other day and we both felt that this is a shame as text mining does not necessarily need a very high level of computer skills - but we still lack good general ICT training for historians...
Text mining for historians
I agree, but history is always about content and context coupled with historiographical techniques. Text mining is part of the picture, but not the whole picture. If a certain historiography requires the engagement with a certain digital corpus, which has almost become the case in the classics community, then the reluctance to use this digital evidence will diminish. In other words it is the underlying research questions that will determine the search for the evidence; digital or otherwise.