Sociology -: A computational approach to sociological explanations

The capacity to explain important elements of social life is central both to the development of sociological theory and to teaching sociology. This research seeks to expand our understanding of sociological explanation through a computational approach. Explanations commonly encountered in introductory sociology texts are used to develop a typology of explanatory forms. A computational strategy that represents sociological knowledge using a combination of frames, semantic networks, and procedural rules is described. It is then demonstrated that this approach can generate the full range of these explanations for all logical combinations of conditions and for the full scope of sociological knowledge. This approach is also shown to be capable of identifying appropriate explanations, assessing the quality of explanations, and generating new insights.