Multiple Uses of Identity Resolution in a Fraud Risk Management System

June 08, 2018 | Identity Resolution, Risk Management

The cost of insurance fraud is staggering. According to the FBI, the total cost of non-health insurance fraud alone is estimated to be more than $40 billion per year, costing the average U.S. family between $400 and $700 per year in the form of increased premiums. A major factor driving fraud is the massive size of the insurance industry itself, with more than 7,000 companies that collect over $1 trillion in premiums each year. An industry of such proportions provides big incentives for committing illegal activities.

The total cost of insurance fraud goes beyond the financial cost of fraudulent insurance claims for insurance companies and increased insurance premiums for consumers. There are additional costs such as regulatory fines, legal and investigative costs, management time, and last but not least, reputational cost.

According to the FBI, the most common types of fraud are premium diversion, fee churning, and asset diversion, but there are many insurance fraud schemes and new ones are constantly appearing. The challenge is how to prevent, detect, and predict fraud.

The Role of Identity Resolution in Insurance Fraud Prevention, Detection, and Prediction

Identity Resolution is a technology designed to detect similar records that differ in possibly multiple, non-obvious ways (e.g., misspellings, spelling variants, nicknames, abbreviations, word ordering, missing or mismatching components) and in any number or combination of record attributes (e.g., name, address, phone number, employer, social security number).  Identity Resolution can play a critical role in many areas of a fraud risk management system, including fraud prevention, detection, and prediction.

For instance, Identity Resolution can be used to screen claimants against third-party data with predictive value (e.g., public records such as bankruptcies, liens, judgements, criminal records, and foreclosures). In the case of health care fraud, providers can be screened against databases of ineligible providers such as those that have been banned by Medicare or are no longer practicing due to retirement or death.

Identity Resolution can also support Social Network Analysis, which may uncover suspicious relationships that may indicate collusion or organized crime rings. For instance, a similar address has been involved in numerous claims, a similar vehicle may have been involved in many accidents with multiple insurance carriers, or the claimant may be related in some way to somebody with a fraud history.

Why Use NetOwl’s Identity Resolution for Insurance Fraud

NetOwl’s Identity Resolution software uses probabilistic machine learning to perform fuzzy matching across records. It combines the evidence from multiple attributes to provide a similarity score. Your organization can set up a high threshold for likely matching records and a lower threshold for human inspection.

NetOwl’s Identity Resolution software is highly accurate. It leverages our industry-leading, award-winning, multicultural, multi-lingual name matching product, NetOwl NameMatcher, to enable sophisticated name matching of various entity types with high accuracy. Its intelligent, probabilistic, name matching rules are derived from large-scale, real-world, multi-ethnicity name variant data.

NetOwl’s Identity Resolution software is also fast and scalable. It works against tens of millions or more entity records with sub-second response times and supports large-scale Big Data applications with hundreds of millions of records.

NetOwl’s Advanced Identity Resolution software can greatly improve your fraud risk management system’s ability to prevent, detect, and predict fraud.