Case Study #1 Combating Insurance Fraud with Data Mining and Analytics By
An assignment submitted in partial fulfillment of the requirement for BKM 400 School of Business Management
February 22, 2013
Insurance companies offer an array of policies to cover a potential loss and/or coverage for valid claims dealing with life, death, health, auto, etc. Unfortunately, fraudulent claims account for a significant portion of claims received by insurers and the cost can exceed billions of dollars. Due to this, insurers have armed themselves with up and coming technically to detect fraud before the company has even made a payment. Companies like Health Care Service Corp. have implemented fraud detection systems that draw their software from IBM and SAS. (Wallace, 2013, p. 239) These systems can detect patterns and trends based on engineering rules created by developers; this has dramatically alleviated the strain felt by insurers to spot suspicious claims. What are some ways that data mining could be used to detect fraud in health insurance claims?
While it is virtually impossible to predict future trends in fraudulent activities, insurers must have the means to become more inventive and resourceful that their criminals. Fraudsters are and can be relentless, especially during times of peril whereas the economy they live is in a recession. However, insurers have more opportunity than ever to recognize fraud and stop it before it occurs by using a combination of approaches and exploiting the advantages of analytic-based techniques. (SAS Institute Inc. World Headquarters, 2013 p.3) To detect fraud in an area such as health insurance claims, data mining software could be used to accesses the unstructured text parsing it to distill meaningful data. This analytical technique analyzes the newly created data to gain a deeper understanding of the claim. (SAS Institute Inc. World Headquarters, 2013 p.7) For example, an...
Please join StudyMode to read the full document