Data processing workflows in the social sciences

摘要

Much of the work of social scientists and government practitioners is consumed by accessing, collating, and analyzing data. This is particularly true in the planning and economic modeling agencies. Unfortunately, there is a severe lack of tools to facilitate this process and much of the integration is done manually by ad-hoc methods. Moreover, raw data are of limited utility. Usually these data are the input to models of more complex phenomena that produce additional data of interest. For example, in our commodity flow domain, we derive truck traffic along specific highway links within a metropolitan area, based on quite far-removed raw (source) data such as employment, imports into and exports out of the region, etc, by using a complex workflow of operations.