Data Mining Solution (DMS) for Child Care Welfare Fraud Detection

About the Program

Category: Information Technology (Best in Category)

Year: 2012

In county government, there is only so much that can be done to combat social welfare fraud. In the arena of child welfare, fraud has become an increasingly large concern that not only reduces the amount of money available to those who are actually in need, but it has also damaged public confidence in child welfare programs. Fraud is frequently perpetrated by organized fraud rings and collusive activities undertaken by providers working with welfare participants. In response, Los Angeles County implemented an innovative data mining solution for child welfare fraud detection. Utilizing an SAS Fraud Framework for Government, the county has incorporated data mining technology with social network, predictive, and forecasting analysis applications. The deployment of predictive fraud detection models has allowed investigators to identify and expedite the review of suspicious cases much earlier than would be the case using traditional investigative techniques and procedures. In identifying cases that match historical patterns of fraudulent activity, investigators can focus on cases with a higher probability of fraud. Consequently, the DMS technology has improved efficiencies in the investigative process since fraud investigators have more time to devote to the review of these high-risk cases. In its first four months of operation, the use of the DMS has resulted in 125 additional referrals for child care fraud investigations. As a result of the Social Network Analysis functionality, the county has uncovered two conspiracy groups consisting of 16 cases and has done so significantly earlier than would have been the case using traditional fraud detection procedures. This early detection significantly reduces the duration of fraudulent activities and thereby generates substantial savings for the county.