Text and data mining (TDM) makes it easier to examine large amounts of data available online and identify interrelationships. With TDM it is possible, for example, to conduct systematic research into the content of large quantities of newspapers. Online solutions such as Voyant often offer limited functionality; in Voyant, for example, one cannot load a large corpus online. Offline applications, such as IBM’s SPSS modeler, often have a steep learning curve.
The Digital Humanities Lab has now launched an online text and data mining application that combines online availability and ease of use with flexibility. The application allows researchers to quickly hook up their own corpus and/or dataset. Successful experiments have already been conducted with large sets of annual reports of financial institutions and a few hundred volumes of the English newspaper the Times, as well as with a collection of Hebrew epigraphs. Currently, the KB newspaper set, the Throne Speeches, and a large set of Spectatorial journals are being added to I-analyzer.