Using Data Mining to Ensure Payment Integrity

Article excerpt

Data mining is the process of analyzing a specific data set with the objective of identifying patterns and establish relationships. Some of the required tools for data mining are importing, analyzing, parsing, matching, summarizing, aging, stratifying, verifying, reporting and documenting data.

Data mining, sometimes referred to as data or knowledge discovery, derives its name from searching for valuable information in a large database, data warehouse or data mart. Such information can be used to establish relationships between variables, and then to validate the findings by applying the detected patterns to new subsets of data. The ultimate goal of data mining is prediction.

Governments spend billions of dollars annually on programs, services and other items. To ensure accountability, governments need to know that these payments are valid and meet compliance requirements. In support of that mandate, payment integrity is essential to enable the governments to manage effectively. For example, federal and state governments spend billions of dollars on health care annually. Improving the accuracy and integrity of government's payments is essential.

Departments of health are committed to ensuring that physicians who provide services to insured persons in their area of jurisdiction receive full payment, within the requirements of the health insurance laws. By using data mining applications, governments can correlate the demographics of patients with critical illnesses to develop better insight on how to identify and treat symptoms and their causes.

Payment integrity is ascertained by using computer query tools and other computer software to conduct routine analysis of the payment requests submitted by the different service providers. It can also be used to conduct investigations into denial of service based either on false rejection or false acceptance. Service providers can also use it to detect duplicate submission of claims.

Software is used to review post-payment claims. Generally, most claims are paid as submitted. However, some claims are checked for validity before payments are made. These checks are general in nature (for example, valid health-ID number, maximum billing of a claim on one day) and cannot validate that the claim submitted is correct or appropriate with the service provided. As a result, post-payment review through the use of software is required.

What can data mining accomplish?

Data mining can perform two basic operations: predicting trends and behaviors and identifying previously unknown patterns. It allows users to analyze data from many different perspectives, categorize the data and summarize the relationships identified. Predictive data mining is most common, and has the most direct business applications. Data mining automates the process of finding predictive information in large databases.

Data mining software is one of a number of analytical tools for analyzing data. Although data mining is a relatively new term, the technology is not. For example, for years, companies have routinely used powerful computers to sift through volumes of point-of-sale scanner data and conduct market research. Another example of a predictive problem is forecasting bankruptcy and other forms of default.

Data mining can also identify previously hidden patterns in a single step. For example, it can analyze retail sales data to discover apparently unrelated products that are often purchased together. One pattern-discovery issue is detecting fraudulent credit card transactions. For example, after using your credit card for some time, a pattern emerges of the typical ways you use your card, such as the places you use it and the amount you spend. If your card is stolen and used fraudulently, the pattern often differs from your own. Data mining tools can distinguish the difference in the two patterns and bring this issue to the attention of the card owner.

Data Mining in Accounting

Mining for accounting demonstrates how companies, financial institutions, insurance companies, tax authorities and other governmental agencies can employ data in accounting, taxation and auditing-related tasks. …