Using data mining to detect crop insurance fraud: is there a role for social scientists?

Abstract

Defines data mining as the extraction of potentially useful information from large databases. Shows how data mining can be applied to detecting anomalous behaviour in American agriculture and thus support the Risk Protection Agency in its compliance mission to detect fraud in crop insurance, using corn as the crop studied and percentage of acres harvested as the key indicator for “proof of concept”. Indicates potential areas of improvement, such as the development of a single data warehouse, and the role of social scientists with knowledge of data analysis and agricultural management. Concludes that data mining could be more effective than the current technique of random selection for investigation of individual entities.