Use Of Data Mining To Detect Fraud Health Insurance Claims

Abstract

The extent, possibility, and complexity of the healthcare industry has attracted widespread fraud that has an impact on the economy. The fraudulent activities not only contribute to the problem of rising healthcare expenditure but also affect the health of patients. The challenge in the current fraud detection systems mostly lies in realizing the burden of the money lost and the unusual behavior areas. Problem Despite putting up various technologies and strategies to fight fraud such as planned, targeted audits, and random audits, whistle blowing, and biometric systems, fraud in claims have continued to be a challenge in most of the health insurance providers in Kenya. Purpose This research tried to analyze the appropriateness of data mining techniques in detecting fraudulent health insurance claims. Method To achieve this, classification models were used to guide the entire knowledge discovery process. Classification algorithms i.e. Naïve Bayes, Decision tree and K-Nearest Neighbor were used to build predictive models. Findings Several experiments were conducted and the resulting models shows that the Naïve Bayes works well in detecting fraud in claims with 91.790% classification accuracy and 74.12% testing hit rate. Value of Study A prototype was developed based on the rules extracted from the Naïve Bayes model which, if adopted, will save costs by detecting fraud as it is committed. Conclusion Fraud detection in health insurance companies, is much needed in developed as well as undeveloped countries so as to help reduce loss of money and in return improve service delivery to patients. Key Words: Healthcare, Health Insurance Claim, Fraud, Naïve Bayes classification, Data Mining