AAAI Publications, 2009 AAAI Fall Symposium Series

Font Size: 
Data Theory, Discourse Mining and Thresholds
David L. Sallach, Jonathan Ozik

Last modified: 2009-10-30

Abstract


The availability of online documents coupled with emergent text mining methods has opened new research horizons. To achieve their potential, mining technologies need to be theoretically focused. We present data theory as a crucial component of text mining, and provide a substantive proto- theory from the synthesis of complex multigames, prototype concepts, and emotio-cognitive orientation fields. We discuss how the data theory presented informs the application of text mining to mining discourse(s) and how, in turn, this allows for modeling across contextual thresholds. Finally, the relationship between discourse mining, data theory, and thresholds is illustrated with an historical example, the events surrounding the 1992 civil war in Tajikistan.

Full Text: PDF