ABSTRACT
The Kansas Event Data System (KEDS) project develops automated natural language processing software, creates specialized event datasets on international political behavior and analyzes these data statistically. Event data – nominal or ordinal codes recording the interactions between political actors as reported in the open press – break down complex activities into a sequence of basic building blocks that can be analyzed statistically. Our machine coding systems, KEDS and TABARI, have been validated against both the textual record and human-coded events (Gerner et al., 1994; Schrodt and Gerner, 1994) and have been used by scholars looking at interactions in Northern Ireland (Thomas, 1999), the Balkans (Goldstein and Pevehouse, 1997; Pevehouse and Goldstein, 1999; Schrodt and Gerner, 2004), the Middle East (Gerner and Schrodt, 1998; Schrodt and Gerner, 1997, 2000; Goldstein et al., 2001; Schrodt, 2006; Brandt and Freeman, 2006; Schrodt and Yilmaz, 2007: Brandt et al., forthcoming), West Africa (Huxtable, 1997), Haiti (Shellman and Stewart, 2006) and the United States (Wood and Peake, 1998).
