New Computer Science Applications in Automated Text Identification and Classification for the Social Sciences

New text-based data resources and tools being developed within the computer and information science communities provide many possibilities for new social science applications. Scholars can envision systematic data of greater consistency, flexibility, coverage, and depth of information than previously possible. However, the diffusion of new techniques from the computer science to the social science community has been slow. Social science applications offer important theoretical challenges to computer scientists as the specific variables of interest to political scientists, sociologists, and others differ from other fields. Social scientists have been developing large databases at a rapid pace in recent years. The large established human-coded databases now in existence provide important means by which we can develop and calibrate new computer-based data development tools. Tighter collaborations across these intellectual communities may thus lead to important theoretical and infrastructure advance in both areas.

This workshop brings together leading computer scientists with political scientists and sociologists with extensive experience in creating large-scale databases. Social scientists will have the opportunity to learn of the latest computer science research relevant to their needs and the computer scientists will learn of the special problems associated with historical research on public policy, social movements, and legislative activities.