Using Computational Techniques to Fill the Gap between Qualitative Data Analysis and Text Analytics

Abstract

The recent mass digitization of text data has led to a need to efficiently and effectively deal with the mountain of textual data that is generated. Digitized text is increasingly in the form of digitized data flows (Brent, 2008). Digitized data flows are non-static streams of generated content – including twitter, electronic news, etc. An oft-cited statistic is that currently 85% of all business data is in the form of text (cited in Hotho, Nürnberger & Paass, 2005). This mountain of data leads us to the question whether the labor-intensive traditional qualitative data analysis techniques are best suited for this large amount of data. Other techniques for dealing with large amounts of data may also be found wanting because those techniques remove the researcher from an immersion in the data. Both dealing with large amounts of data and allowing immersion in data are clearly desired features of any text analysis system.