Solutions for Fraud Detection: Medical Insurance Fraud, Medicare Fraud 

STATISTICA Data Miner solutions provide a large selection of the most powerful algorithms and methods for "anomaly detection", or "fraud detection". Fraud is costly in any business, and can particularly affect the bottom line for products offered by insurance companies (see also STATISTICA Data Miner Predictive Modeling Solutions for the Insurance Industry). There are many approaches available through STATISTICA Data Miner based solutions that are effective at identifying fraud components in various insurance domains (see Fraud Detection in the Online Electronics Textbook); however, it is particularly difficult to detect fraud in the complex system of reimbursements for medical services.

Health insurance fraud and Medicare fraud can be extremely costly, and difficult to detect. Because typical datasets (data bases) are extremely large (both with respect to the number of records and fields), and because the coding of medical diagnoses and services is complex and detailed, finding patterns associated with a high likelihood of fraud (or recovery) is particularly difficult. At the same time, given the large amount of money that is dispersed by, for example, Medicare, even the smallest increase in fraud translates into very large amounts of money, and "disruption" to health care delivery given the small margins and tight budgets of many health care providers.

STATISTICA Enterprise / Data Miner based solutions provide the right mix of technology and capabilities to address these challenges: STATISTICA Data Miner offers cutting edge predictive modeling, clustering, and anomaly detection methods in a software architecture that is highly scalable to support multithreaded and parallel processing of data from multiple data sources. STATISTICA solutions can be deployed for ad-hoc modeling and analyses, automated screening in batch mode (e.g., end-of-day processing of claims), or operate real-time supporting existing IT assets and systems to add critical ROI by identifying new patterns of likely fraudulent activities or unjustified reimbursements.